Intel Xeon Processor 5500 Series (Nehalem-EP) Performance Package

Intel Xeon Processor 5500 Series (Nehalem-EP) ( ) Performance Package ® ® Bob Burroughs Server Platforms Group p Marketing g Approved for external u...
Author: Ashlyn Dean
3 downloads 4 Views 12MB Size
Intel Xeon Processor 5500 Series (Nehalem-EP) ( ) Performance Package ®

®

Bob Burroughs Server Platforms Group p Marketing g Approved for external usage Rev 1.45 – June 11, 2009

1

Legal Disclaimers Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/benchmark_limitations.htm or call (U.S.) 1-800-628-8686 or 1-916-356-3104. Intel does not control or audit the design or implementation of third party benchmarks or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmarks are reported and confirm whether the referenced benchmarks are accurate and reflect performance of systems available for purchase. Relative R l ti performance f i calculated is l l t d by b assigning i i a baseline b li value l off 1.0 1 0 to t one benchmark b h k result, lt and d then th dividing di idi the th actual t l benchmark b h k resultlt for f the th baseline b li platform into each of the specific benchmark results of each of the other platforms, and assigning them a relative performance number that correlates with the performance improvements reported. SPEC, SPECint, SPECfp, SPECrate. SPECpower, SPECjAppServer, SPECjbb, SPECjvm, SPECWeb, SPECompM, SPECompL, SPEC MPI, are trademarks of the Standard Performance Evaluation Corporation. See http://www.spec.org for more information. TPC-C, TPC-H, TPC-E are trademarks of the Transaction Processing Council. See http://www.tpc.org for more information. Intel® Virtualization Technology requires a computer system with an enabled Intel® processor, BIOS, virtual machine monitor (VMM) and, for some uses, certain platform software enabled for it. Functionality, performance or other benefits will vary depending on hardware and software configurations and may require a BIOS update. Software applications may not be compatible with all operating systems. Please check with your application vendor. Hyper-Threading Technology requires a computer system with a processor supporting HT Technology and an HT Technology-enabled chipset, BIOS and operating system. Performance will vary depending on the specific hardware and software you use. For more information including details on which processors support HT Technology see here Technology, Intel® Turbo Boost Technology requires a Platform with a processor with Intel Turbo Boost Technology capability. Intel Turbo Boost Technology performance varies depending on hardware, software and overall system configuration. Check with your platform manufacturer on whether your system delivers Intel Turbo Boost Technology. For more information, see http://www.intel.com/technology/turboboost ” Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor series, not across different processor sequences. See http://www.intel.com/products/processor_number sequences http://www intel com/products/processor number for details. details Intel products are not intended for use in medical, medical life saving, saving life sustaining, sustaining critical control or safety systems, or in nuclear facility applications. All dates and products specified are for planning purposes only and are subject to change without notice * Other names and brands may be claimed as the property of others. Copyright © 2009 Intel Corporation. All rights reserved. Intel, the Intel logo, Xeon and Intel Core are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. All dates and products specified are for planning purposes only and are subject to change without notice

2

Table of Contents

3

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® 5500 Processor • Intelligent Performance • Adaptable Energy Efficiency • Flexible Virtualization Enabled by – – – – – – – –

Intel® Microarchitecture Nehalem 45 nm Hi-K Quad Core processor Intel® Turbo Boost Technology Intel® l Hyper-Threading h d Technology h l Enhanced Virtualization Technology Integrated Power Gates Automated Low-Power States Intel® Node Manager

A new generation of Intelligent Server Processors

4

Intel® Xeon® 5500 Platform New Memory Subsystem NEW! Intel® QuickPath Interconnect Intel® Data Center Manager

Intel® 5520 NEW! Chipset

Intel® Intelligent Power Technology

NEW!

Intel® Node Manager

New I/O Subsystem

PCI Express* 2.0

NEW! Intel® X25-E SSDs

ICH 9/10

Intel® 82599 10GbE Controller

Massive Platform level innovations

5

Product Options Intel Xeon Processor 5500 Series ®

CPU Number

®

Usage

Link Speed

Max Mem Speed

8M

64 6.4

1333

4

8M

5.86

1066

GT/s

MHz

2-4

4M

4.8

800

Cores

Cache

4

Turbo† HT

95W

X5570 (2.93 GHz)

Ad Advanced d

X5560 (2.80 GHz)

GT/s

MHz

+3

9

+2

9

No Turbo

No HT

X5550 (2.66 GHz) 80W

St d d Standard

E5540 (2.53 GHz) E5530 (2.40 GHz) 80W

60W

E5520 (2.26 GHz)

L5520

E5506 (2.13 GHz)

L5506

80W

E5504 (2.00 GHz) E5502 (1.86 (1 86 GHz)

Low Power Basic

GT/s

MHz

Dual Core

Value add features available on higher end processors †

6

Max Turbo Boost frequency based on number of 133 MHz increments above base freq (+2 = 0.266 GHz, +3 = 0.400 GHz)

Table of Contents

7

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® 5500 Performance Publications SPECint*_rate_base2006

SPECpower*_ssj2008

SPECfp*_rate_base2006

241 score

1977 ssj_ops/watt ssj_ops/watt IBM J9* JVM

197 score

(+72%)

(+74%)

(+128%)

SPECjAppServer*2004 SPECjAppServer *2004

TPC*--C TPC*

3,975 JOPS Oracle WebLogic* Server

631,766 tpmC Oracle 11g* database

VMmark** VMmark

TPC*--E TPC*

SPECWeb*2005 SPECWeb *2005

24.24 @17 tiles VMware* ESX 4 VMware 4.0 0

800 tpsE Microsoft SQL Server* Server 2008

75023 score Rock Web Web* Server

Fluent* 12.0 benchmark

SPECjbb*2005 SPECjbb *2005

SPECapc* for Maya 6.5

Geo mean of 6 ANSYS FLUENT*

604,417 BOPS IBM J9* JVM

7.70 score Autodesk* Maya

(+93%)

(+165%)

(+127%) ( )

(+130%)

(+152%)

((+64%))

SAP--SD* 2 SAP 2--Tier

(+103%)

5,100 SD Users SAP* ERP 6.0/IBM DB2*

(+150%)

((+87%))

Over 30 New 2S Server and Workstation World Records! Percentage gains shown are based on comparison to Xeon 5400 series; Performance results based on published/submitted results as of April 27, 2009. Platform configuration details are available at http://www.intel.com/performance/server/xeon/summary.htm *Other names and brands may be claimed as the property of others

8

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations

Intel® Xeon® Processor 5500 series based Server platforms Server Performance comparison to Xeon 5400 Series

Relative Performance Higher is better

Xeon 5400 series Server Side Java

I t Integer

Energy Efficiency

Java Apps

App Server

ERP

Floating point

D t b Database

D t b Database

W b Web

Virtualizat ion

Source: Published/submitted/approved results June 1,2009. See backup for additional details

Leadership on key server benchmarks

9

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms HPC Performance comparison to Xeon 5400 Series

Relative Performance Higher is better

Xeon 5400 series W th Weather

FEA

FEA

CFD

CFD

E Energy

O Open MP

E Energy

O Open MP

W th Weather

E Energy

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

Exceptional gains on HPC applications

10

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to AMD Opteron 2300 Series “Shanghai” Relative Performance Higher is better

Baseline(0%)– Best Quad-Core AMD Opteron* 23xx published result (“Shanghai”)

Source: Published/submitted/approved results June 9, 2009. See backup for additional details

Significant Leadership in Server, HPC, Virtualization and Energy Efficiency benchmarks 11

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Server Performance comparison to six-core AMD Istanbul

Relative Performance Higher is better

Six-Core Istanbul 2.6 GHz

Server Side Java

Integer

Floating point p

Virtualizat ion

ERP

Source: Published/submitted/approved results June 11,2009. See backup for additional details

Xeon 5500 Series delivers strong leadership over Istanbul

12

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 Series based Platforms ISV Application Performance

ISV Application

Xeon 5500 vs Xeon 5400

ISV Application

Xeon 5500 vs Xeon 5400

Ansys* CFX11* – CFD Simulation

+88%

Kingsoft* JXIII Online* – Game Server

+98%

ERDAS* ERMapper Suite*

+69%

Mediaware* Instream* – Video Conv.

+73%

ESI Group* PAM-CRASH*

+50%

Neowiz* Pmang* – Game Portal

+100%

ExitGames* Neutron ExitGames Neutron* – Service Platform

+80%

Neusoft* – Healthcare Neusoft

+131%

Epic* – EMR Solution

+82%

Neusoft* – Telecom BSS VMware*

+115%

Giant* Juren* – Online Game

+160%

NHN Corp* Cubrid* – Internet DB

+44%

IBM* DB2* 9.5 – TPoX XML

+60%

QlikTech* QlikView* – BI

+36%

IBM* Informix IBM Informix* Dynamic Server

+84% 84%

SAS* Forecast Server* SAS Server

+80% 80%

IBM* solidDB* – In Memory DB

+87%

SAP* NetWeaver* – BI

+51%

Image Analyzer * – Image Scanner

+100%

SAP* ECC 6.0* – ERP Workload

+71%

Infowrap* – Small Data Set

+129%

Schlumberger* Eclipse300*

+213%

p – Weather Forecast Infowrap*

+155%

SunGard* BancWare Focus ALM*

+38%

Intense* IECCM* – Output Mgmt.

+150%

Supcon* – APC Intel. Sensor

+191%

Intersystems* Cache* – EMR

+63%

TongTech* – Middleware

+95%

Kingdee* APUSIC* – Middleware

+93%

UFIDA* NC* – ERP Solution

+230%

g EAS* – ERP Kingdee*

+142%

UFIDA* Online – SaaS Hosting g

+237%

Kingdom* – Stock Transaction

+141%

Vital Images* – Brain Perfusion 4DCT*

+77%

Exceptional gains (1.6x to 3x) on ISV applications Source: Results measured and approved by Intel and ISV partners on pre-production platforms. March 30, 2009.

13

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Performance Comparison over multiple generations

R Relative Performa P ance

Relative Performance Higher is better

Intel® Xeon® 5500

Performance gains represent a blend (GEO mean) of five common 2-socket workload types across a range of typical usages

Intel® Xeon® 5300

Intel® Xeon® 5400

Intel® Xeon® 5100

Intel Single Single-Core Core

Intel Multi Multi-Core Core

Compelling performance gains Enables unprecedented opportunities for business Source: Published/submitted/measured results March 30, 2009. Each bar represents the geo mean of best published results on five industry standard benchmarks – SPECint*_rate_base2006, SPECfp*_rate_base2006, SPECjbb*2005, TPC*-C and SAP-SD*.

14

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Server Refresh opportunity

2005

2009 Performance Refresh 1:1

Up to

Intel®

Xeon®

184 5500 Based Servers

9x Performance

18% Annual Energy Costs Estimated Reduction

– OR – 90% % Annual Energy gy

Efficiency y Refresh 9:1

184 Intel® Xeon® Single Core Servers

Costs Estimated Reduction Intel®

Xeon®

21 5500 Based Servers

As Low as

8 Month Payback

Source: Intel estimates as of Nov 2008. Performance comparison using SPECjbb2005 bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup.

15

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

16

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms Integer Throughput performance on SPECint*_rate_base2006

Higher is better

Enhanced General Purpose Computing



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by ASUSTex on ASUS Z8PED12X server motherboard based platform



Intel Compiler 11 11.0 0 based binaries



SLES 10 SP2 OS

+72%

Benchmark notes: • SPECint*_rate_base2006 benchmark measures Server’s capacity to perform a number of parallel Integer tasks Source: Published/submitted/approved results June 9,2009. See backup for additional details

Xeon X5570 outperforms six-core Istanbul by 1.49x Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2360 – Quad-Core AMD Opteron* Model 2360SE (“Barcelona 2.50GHz”) Xeon 5470 – Intel® Xeon® Processor X5470 (“Harpertown 3.33GHz”, Quad-Core) Opteron 2389 – Quad-Core AMD Opteron* Model 2389 (“Shanghai 2.90GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 17 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point Throughput performance on SPECfp*_rate_base2006

Higher is better

Enhanced Scientific Computing



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by Fujitsu on PRIMERGY TX200 Server



Intel Compiler 11.0 based binaries



SLES 10 SP2 OS

+125%

Benchmark notes: • SPECfp*_rate_base2006 SPECf * t b 2006 benchmark measures Server’s capacity to perform a number of parallel Floating point tasks Source: Published/submitted/approved results June 9,2009. See backup for additional details

125% gain on SPECfp_rate* demonstrates the dramatic increase platform bandwidth, 1.52x six-core Istanbul p performance in p Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2360 – Quad-Core AMD Opteron* Model 2360SE (“Barcelona 2.50GHz”) Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Opteron 2389 – Quad-Core AMD Opteron* Model 2389SE (“Shanghai 2.90GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 18 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer “SPEED” performance on SPECint*_base2006

Higher is better

Enhanced General Purpose Computing for Single-threaded Apps



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by ASUSTeK Computer Inc on RS700-E6 server



Intel Compiler 11.0 based binaries



SLES 10 SP2 OS

+21%

Benchmark notes: • SPECint*_base2006 SPECi t* b 2006 benchmark measures Server’s capacity to perform a Single Integer task Source: Published/submitted/approved results June 9,2009. See backup for additional details

Turbo mode enables strong gains on Integer Single Task performance Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Xeon 5470 – Intel® Xeon® Processor X5470 (“Harpertown 3.33GHz”, Quad-Core) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 19 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point “SPEED” performance on SPECfp*_base2006

Higher is better

Enhanced Scientific Computing for Singlethreaded Apps



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by HITACHI on Blade Symphony BS2000



Intel Compiler 11.0 based binaries



SLES 10 SP2 OS

+48%

Benchmark notes: • SPECfp*_base2006 SPECf * b 2006 benchmark measures Server’s to perform a single Floating point task

Source: Published/submitted/approved results June 9,2009. See backup for additional details

Up to 45% improvement on Floating Point Single Task performance Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2360 – Quad-Core AMD Opteron* Model 2360SE (“Barcelona 2.50GHz”) Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 20 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java performance on SPECjbb*2005

Oracle JRockit* JVM

Java HotSpot* JVM

IBM J9* JVM

Faster Java Servers Higher is better Business Operations Per Hour

+51%

+64%

+75%

No Published Result

Source: Published/submitted/approved results June 9,2009. See backup for additional details

Leadership on major JVMs Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2434 – Six-Core AMD Opteron* (“Istanbul 2.60GHz”) Xeon 5470 – Intel® Xeon® Processor X5470 (“Harpertown 3.33GHz”, Quad-Core) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 21 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Database performance on TPC*-E

Higher is better

Faster database servers for Online Transactions



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by Fujitsu on PRIMERGY RX300 G5 Server



Microsoft SQL Server 2008 database

+152%

No Published result

$1897/ tpsE

$524/ tpsE

$344/ tpsE

Benchmark notes: • TPC*-E measures database server performance in a brokerage firm environment • Metric: Transactions Per Second (tpsE)

Source: Published/submitted/approved results June 9,2009. See backup for additional details

152% higher performance on TPC-E Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Xeon 5460 – Intel® Xeon® Processor X5460 (“Harpertown 3.16GHz”, Quad-Core)

22

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Database performance on TPC*-C



Comparison basis: X86 architecture based DP Server platforms



Leading results published b H by Hewlett-Packard* l tt P k d* on HP ProLiant* DL370 G6 System using on Oracle* 11g database



Xeon 5570 with 144GB memory Xeon 5160 and 5470 with 64GB memory

Higher is better

Faster database servers for Online Transactions

+125% 125%

No Published result

$2.40/ tpmC

$2.93/ tpmC

$1.38/ tpmC

$1.44/ tpmC

$1.08/ tpmC

Source: Published/submitted/approved results June 9,2009. See backup for additional details



Benchmark notes: • TPC*-C measures database server performance in an On-line transaction processing environment • Metric: Transactions Per Minute (tpmC)

TPC-C leadership on Oracle database Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Xeon 5460 – Intel® Xeon® Processor X5460 (“Harpertown 3.16GHz”, Quad-Core) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 23 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Web Server performance on SPECWeb*2005



Comparison basis: X86 architecture based DP Server platforms



Leading result on Xeon X5570 published by HP on DL380 G6 Server

• •

Rock Web Server 1.4.7 Red Hat Linux 5.3

Higher is better

More HTTP Requests handled Per Second

+150%

Source: Published/submitted/approved results June 9,2009. See backup for additional details

Benchmark notes: • Measures performance of Web Servers • Benchmark consists of 3 workloads - Banking (HTTPS), E-commerce (HTTP and HTTPS), and Support (HTTP) • Final result is the geo mean of scores of the three workloads

Dramatic increase in SPECWeb2005 performance Xeon 3.80 – Intel® Xeon® Processor 3.80 2M L2 (“Irwindale 3.80GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2360 – Quad-Core AMD Opteron* Model 2360SE (“Barcelona 2.50GHz”) Xeon 5460 – Intel® Xeon® Processor X5460 (“Harpertown 3.16GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 24 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Enterprise Resource Planning performance on SAP-SD* 2-Tier

Higher is better

Non-unicode/Old benchmark

Unicode/New benchmark

More SAP Users supported by Server



Comparison basis: X86 architecture based DP Server platforms



Leading g result on Xeon X5570 published by IBM on System x3650 M2 Server



All results use nonunicode format except one

SAP ERP 6.0 Enhancement pack 4 Unicode

SAP ERP 6.0 (2005) nonUnicode

+103%

Source: Published/submitted/approved results June 2, 2009. See backup for additional details

Benchmark notes: • SAP Sales and Distribution (SD) benchmark • Forms the basis of SAP sizing forecasts • Metric: Number of SD users

Xeon X5570 results of 5100 were published in Dec 2008 on the prior version of benchmark and using non non-unicode unicode mode mode. Xeon X5570 score of 3700 was using the revised benchmark version using uni-code mode. These results are not apples-to-apples

Up to 103% higher performance than 5400 series p to 1.57x six-core Istanbul Performance Up Xeon 3.60 – Intel® Xeon® Processor 3.60 2M L2 (“Irwindale 3.60GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Xeon 5470 – Intel® Xeon® Processor X5470 (“Harpertown 3.33GHz”, Quad-Core) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 25 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java Application and Application Server performance

+93%

Higher is better Java Operations Per Second

Higher is better Operations/m

+77%

No Published result

No Published result

Dell PowerEdge 710 platform Oracle Weblogic Application Server, Oracle Jrockit JVM, Oracle 11g Database

Sun Blade X6270 platform OpenSolaris 2008.11 HotSpot JVM - JDK 1.6.0 1.6.0_14 14

Solid gains on Java benchmarks Xeon 5470 – Intel® Xeon® Processor X5470 (“Harpertown 3.33GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results June 1,2009. See backup for additional details

26

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

27

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficient performance on SPECpower*_ssj2008



C Comparison basis: X86 86 architecture based DP Server platforms



Leading result by IBM on IBM System x 3650 M2 server platform on Xeon X X5570 • IBM J9* JVM • Microsoft Windows Server 2008 Enterprise*

Higher is better ssj_ops/watt

More Energy efficient servers

+74%

Intel 5000P chipset “Blackford”

Intel 5100 chipset “San Clemente”

Source: Published/submitted/approved results June 9,2009. See backup for additional details

Perf/Watt Perf /Watt

Benchmark notes: • Measures energy efficiency of volume servers • First industry standard bmk to measure power consumption in relation to performance across a g workload ((11 different “graduated” load levels -100%, 90%, 80% down to 0%) • Measures platform power - AC watts at the wall • Metric: Overall ssj_ops / watt (sum perf p points divided by y of the 11 p sum of the 11 power points)

Xeon 5570 delivers all-time high SPECpower*_ssj2008 score Xeon 3.60 – Intel® Xeon® Processor 3.60 1M L2 (“Nocona 3.60GHz”, Single-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Opteron 2376 – Quad-Core AMD Opteron* Model 2376HE (“Shanghai 2.30GHz”) Xeon 5470 – Intel® Xeon® Processor L5430 (“Harpertown 2.66GHz”, Quad-Core) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 28 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficient performance on SPECint*_rate and SPECfp*_rate

Perf/Watt

Higher is better

Energy Efficient Integer Performance

+85%

Energy Efficient Floating point Performance

Intel 5000P chipset “Blackford”

Source: Published results at www.principledtechnologies.com as of March 30, 2009. See backup for details

+120%

Intel 5400 chipset “Seaburg”

Relative Performance. Higher is better

Exceptional increase in Energy Efficient Performance Xeon 3.80 – Intel® Xeon® Processor 3.60 2M L2 (“Irwindale 3.60GHz”, Single-Core) Xeon 5472 – Intel® Xeon® Processor E5472 (“Harpertown 3.0GHz”, Quad-Core) Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5450 – Intel® Xeon® Processor E5450 (“Harpertown 3.0GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 29 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Performance vs Power - load line comparison of WS 2008 SP2 vs WS 2003 SP2 Perf/Watt



Compares the performance two different versions of OS



Platform details SuperMicro Server Preproduction platform

– –

PPW gains at each load level



• WS08 vs WS03 Out-of-box measurement highlights



Xeon X5570 processor 12GB (6x2GB) memory

OS versions compared p



Microsoft Windows Server 2003 SP2



Microsoft Windows Server 2008 SP2

SPECpower*_ssj2008 used for comparison

Joint collaboration between Intel & Microsoft WS08 harnesses Xeon 5500 series (“Nehalem-EP”) energy efficiency features - P & C state innovations Average Performance Per Watt improvement of 12% in the 10-50% utilization range – The sweet spot of Server Platform Over 6% improvements @ idle; Over 7% improvement in overall SPECpower score

Improved Power management capabilities in Win Server 2008 SP2 enables higher energy efficiency compared to Win Server 2003 SP2 Source: Intel and Microsoft approved result -March 30, 2009.

30

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

31

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms

Virtualization performance on VMware ESX* using VMmark* benchmark Higher is better

More VMs per Server

+156% 156%

+161% 161%

+162% 162%

+164% 164%

+165% 165%

Benchmark Description: VMmark* is designed as a tile-based benchmark consisting of a diverse set of workloads commonly found in the datacenter, The workloads comprising each tile are run simultaneously in separate virtual machines at load levels that are typical of virtualized environments. The performance of each workload is measured and then combined with the other workloads to form the score for the individual tile. Multiple tiles can be run simultaneously to increase the overall score. A tile is a collection of six diverse workloads concurrently executing specific software software. Running on one of two separate operating systems (Windows or Linux) Linux), each workload runs in its own virtual machine and executes specific applications. Included in a single tile are a web server, file server, mail server, database, Java server, as well as an idle machine. Results are posted at http://www.vmware.com/products/vmmark/results.html

Xeon 5500 series delivers Up to 165% gain on VMmark & outperforms six six-core core Istanbul by 1.56x 1 56x Source: Published/submitted/approved results as of June 8, 2009. See backup for additional details

32

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Legacy Database Server Consolidation under Virtualization with Microsoft Hyper-v



Description



The chart illustrates consolidation of one or more legacy database servers on to a single virtualized server platform. The database server used is Microsoft SQL Server* 2005 and the virtualization software used was Microsoft Windows Server* 2008 with Hyper-V. The legacy server, which was not virtualized, forms the baseline platform. On the consolidation server, virtual machine (VM) partitions of size comparable to the baseline platform l tf were created. t d The Th maximum i number b off such h concurrentt VMs that can be supported defines the consolidation ratio. The application used was DVD Store Version 2 (DS2) test tool. DS2 is an open-source simulation of an online e-commerce DVD store, where customers log in, browse, and order products. The main DS2 metric is orders per minute, or OPM. Comparison basis: X86 architecture based DP Server platforms



Higher is better Orders Per Minute (OPM)

Baseline platform details: • Single Core Intel Xeon processor 3.60GHz (‘Irwindale 3.60”). Launched in 2005



• Microsoft SQLServer 2005 database • Microsoft Windows Server 2003 SP2 • 4GB memory Consolidated platform • Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V.



• 64GB memory for Xeon E5450 and Opteron 2384 • 72GB memory for Xeon X5570 Additional details:



Note that some VMs show more OPM than the baseline. The criteria was at least baseline score

Native

Virtualized

Virtualized

Virtualized

CPU @

77%

99%

99%

100%

Avg OPM

3632

4272

5108

3744

5

8

5

# VM VMs

Source: Result published at www.principledtechnologies.com as of March 30, 2009. See backup for details

Xeon X5570 demonstrates excellent consolidation opportunity for Legacy Database servers (up to eight servers)

33

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Legacy Mail Server Consolidation under Virtualization with Microsoft Hyper-v



Description



The chart illustrates consolidation of one or more legacy mail servers on to a single virtualized server platform. The mail server used is Microsoft Exchange* Server 2003 and the virtualization software used was Microsoft Windows Server* 2008 with HyperV. The legacy server, which was not virtualized, forms the baseline platform. Microsoft Exchange Load Simulator’s MAPI Messaging Benchmark 3 (MMB3) workload was used as the load simulator. i l t W Workload kl d consisted i t d off 8 8,000 000 users. During D i the th ttests, t LoadSim performs tasks to simulate a standard user generating mail activity. When the workload finishes, LoadSim reports the response time, which is the number of seconds necessary to complete the task. Transaction latency was defined to be less than one second. On the consolidation server, virtual machine (VM) partitions of size comparable to the baseline platform were created. The maximum number of such concurrent VMs that can be supported defines the consolidation ratio.



Higher is better Number of Exchange users

Baseline platform details: • Single Core Intel Xeon processor 3.60GHz (‘Irwindale 3.60”). Launched in 2005



• Microsoft Exchange Server 2003 • Microsoft Windows Server 2003 SP2 • 4GB memory Consolidated platform • Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V.



• 64GB memory for Xeon E5450 and Opteron 2384 • 72GB memoryy for Xeon X5570 Additional details:



Note that Xeon X5570 has idle cpu. Adding more VMs resulted in transaction latency of >1 sec.

Native

Virtualized

Virtualized

Virtualized

CPU @

73%

93%

75%

90%

Workload

8000 users

8000 users/VM

8000 users/VM

8000 users/VM

5

9

5

# VM VMs

Source: Result published at www.principledtechnologies.com as of March 30, 2009. See backup for details

Xeon X5570 demonstrates excellent consolidation opportunity for Legacy Mail servers (up to nine servers)

34

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

35

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms

Open MP performance on SPECOMP*2001 – Medium and Large Workloads

Higher is better

+154%

Sun Blade X6275

Higher is better

+192%

Sun Blade X6275

Source: Published/submitted/approved results June 9 2009. See backup for additional details

Huge improvements on bandwidth intensive, HPC centric, SPEC OMP* 2001 benchmarks Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Xeon 5472 – Intel® Xeon® Processor E5472 (“Harpertown 3.0GHz”, Quad-Core) Opteron 2360 – Quad-Core AMD Opteron* Model 2360SE (“Barcelona 2.50GHz”) Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 36 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Computation Fluid Dynamics Performance using ANSYS* FLUENT*

• Comparison Details: –

Comparison based on published results at www.fluent.com/software/fluent/fl6bench/fl6b ench_6.4.x/index.htm as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Higher is better

• Xeon X5570 performance result published by SGI on SGI Altix ICE 8200EX* server platform • Benchmark details: –

ANSYS FLUENT12 is a commercial computational fluid dynamics (CFD) package and is a product of ANSYS Inc. It has broad physical modeling capabilities well suited to a variety y of industrial applications. pp Fluent provides a set of standardized benchmarks covering a range of fluid flow simulations and software features. Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchma ks Performance benchmarks. Pe fo mance metric met ic is called Solver rating

127% increase in performance compared to Xeon 5400 series (Based on Geo mean result) Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

37

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Crash Simulation Analysis using LS-DYNA*

• Comparison Details: –

Comparison based on published results at http://www.topcrunch.org/ as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Relative Performance Higher is better

• Xeon X5570 performance result published by Intel on SuperMicro* X8DTN+ platform • Benchmark details: –

LS DYNA is a general purpose transient dynamic LS-DYNA finite element program capable of simulating complex real world problems, for use in various industries, including Automobile Design, Aerospace, Manufacturing, and Bioengineering. Benchmark data sets and cluster performance quotations are available from the independent web site www.topcrunch.org. These benchmarks are associated with structural dynamics (LSDYNA) address domain decomposition, message passing, load balancing, and dynamic memory allocation in automotive crash safety analysis. The chart shows single node benchmark performance comparisons of automotive crash simulation for single vehicle barrier crash, 3 vehicle rear end crash, and 2 vehicle head-on crash (neon_refined_revised, 3 vehicle collision, car2car).

Nearly y 2X g gain on LS-DYNA Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

38

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Multiphysics Finite Element Analysis using ANSYS* Mechanical*

• Comparison Details: –

Comparison based on published/submitted results to www.ansys.com/services/hardwaresupport-db.htm as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quadcore servers.

Relative Performance Higher is better

• Xeon X5570 performance result published by Intel on SuperMicro* X8DTN+ platform • Benchmark details: –

ANSYS 12* software is a comprehensive multiphysics tool combining structural, thermal, fluids, acoustic and electromagnetic simulation capabilities in a single engineering software solution. It's comprehensive range of physical models can be applied to simulation-based product development in a broad range of industries and applications. The "shared" benchmark is a suite of 8 workloads and the "distributed" benchmark is a suite of 7 workloads and they cover a representative set of structural analysis solvers and analysis types.

Up to 115% gain on ANSYS Mechanical 12.0 Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

39

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Unified Finite Element Analysis using Abaqus*

• Comparison Details: –

Comparison based on published results to www.simulia.com/support/v68/v68_performance. php as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Relative Performance Higher is better

• Xeon X5570 performance result published by Intel on SuperMicro* X8DTN+ platform • Benchmark details: –

The Abaqus Unified FEA product suite offers powerful and complete solutions for both routine and sophisticated engineering problems covering a vast spectrum of industrial applications. Abaqus/Standard benchmarks include linear statics, nonlinear statics, and natural frequency extraction workloads. Abaqus/Explicit benchmarks include workloads modeling high-speed dynamic impact events and quasi-static events with complicated contact conditions. Benchmark for Abaqas/Standard consists of a suite of 10 workloads and Abaqas/Explicit consists of a suite of 6 workloads. Geo mean of each these workload groups used for comparison. comparison

Up to 70% increase on Abaqus benchmarks Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core) Opteron 2384 – Quad-Core AMD Opteron* Model 2384 (“Shanghai 2.70GHz”) Source: Published/submitted/approved results March 30, 2009. See backup for additional details

40

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Computational Fluid Dynamics Analysis using Star-CD*

• Comparison Details: –

Comparison based on published results to www.cd-adapco.com/products/STARCD/performance/406/index.html as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Relative Performance Higher is better

• Xeon X5570 performance result published by SuperMicro* X8DTN+ platform • Benchmark details: –

– –

STAR-CD* provides a platform for industrial CFD simulation. Going beyond just a CFD code, the latest release, STAR-CD V4, introduces the capability to perform structural analysis calculations using a methodology based upon its CFD solver technology, a comprehensive solution for flow, thermal and stress simulation has been available in a single general-purpose commercial finite-volume code. STAR-CD V4 is fully polyhedra enabled and by using the latest polyhedra-solver polyhedra solver technology, STAR-CD V4 delivers significant benefits in speed, robustness and usability. Workloads for Star-CD (A-class, C-class): The key workloads for Star-CD benchmark are AClass and C-Class. The A-class (resp. C-class) workloads simulate the turbulent, turbulent steady-state flow around an A-Class (resp. C-class) Mercedes* car. Both workloads use the Conjugate Gradient Solver. A-class uses a trimmed hexahedral mesh made of 6 million cells while C-class uses a polyhedral mesh of 10 million cells.

Up tto 117% iincrease on St U Star-CD CD benchmarks (Single Precision)

Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 41 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Reservoir Simulation Using Schlumberger Eclipse*

• Comparison Details: –

Comparison based on ISV approved results at as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Relative Performance Higher is better

• Xeon X5570 performance result published by Intel on SuperMicro* X8DTN+ platform • Benchmark details: –

Eclipse* has been the leading commercial reservoir simulation software for over 25 years because of its breadth of capabilities, parallel scalability, utility computing, and unmatched platform coverage. Eclipse Version 2008.1 continues to deliver accurate, high performance solutions for all of today's complex reservoir systems. The two workloads shown demonstrate Eclipse capabilities in handling black oil and compositional simulations. Eclipse 100* is a one million cell black-oil model, and Eclipse 300* is a two million cell compositional model.

Exceptional gains (150%) gain on Eclipse p Benchmarks Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core)

Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

42

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Reservoir Simulation using CMG* IMEX*

• Comparison Details: –

Comparison based on ISV approved results at as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Higher is better Relative Performance Higher is better

• Xeon X5570 performance result published by Intel on SuperMicro* X8DTN+ platform

+139%

• Benchmark details: –

IMEX, CMG's full featured Black Oil simulator, models the flow of three phase fluids in gas, gas gas-water gas water, oiloil water, and oil-water-gas reservoirs. It models in one, two, or three dimensions, including complex heterogeneous faulted structures. It also models primary, secondary, and pseudo-miscible and polymer injection processes in naturally or hydraulically fractured reservoirs. The workload is a Black-oil Water Flood simulation with CMG IMEX app. Simulated 545K Cells with 50 wells simulated for 15 years.

Exceptional scaling on CMG IMEX Xeon 5160 – Intel® Xeon® Processor 5160 (“Woodcrest 3.0GHz”, Dual-Core)

Relative Performance. Higher is better Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Xeon 5570 – Intel® Xeon® Processor X5570 (“Nehalem-EP 2.93GHz”, Quad-Core)

Source: Published/submitted/approved results March 30, 2009. See backup for additional details

43

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Reservoir Simulation using Landmark Nexus*

• Comparison Details: –

Comparison based on ISV approved results as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.

Higher is better Relative Performance Higher is better

• Xeon X5560 performance result published by Intel on SuperMicro* X8DTN+ platform

+196%

• Benchmark details: –

Landmark's Nexus* is a major reservoir simulation software that enables fully implicit, fully coupled surfaceto-subsurface simulation at unprecedented speed and accuracy. SPE10* is an industry-standard reservoir model derived from an actual oil reservoir in the North Sea.

196% improvement on Landmark Nexus*

Relative Performance. Higher is better

Xeon 5482 – Intel® Xeon® Processor X5482 (“Harpertown 3.20GHz”, Quad-Core) Xeon 5560 – Intel® Xeon® Processor X5560 (“Nehalem-EP 2.80GHz”, Quad-Core) Source: Published/submitted/approved results March 30, 2009. See backup for additional details

44

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms SPEC MPI* 2007 performance on a multi-node cluster

Higher is better



Key Details Cluster configuration, Comparison on equal number of Nodes, Each node contains two quad-core cpus. 1 thread/core Result published by SGI on SGI Altix ICE 8200EX* server based cluster, Comparison based on “base” score results All data based on published results at http://www.spec.org/mpi2007/results/mpi2007.html as of March 30, 2009



Benchmark notes Benchmark suite for evaluating MPI-parallel, floating point, compute intensive performance for cluster and SMP hardware. Developed from native Message Passing Interface (MPI) parallel end-user applications

Excellent cluster scaling g on SPEC MPI 2007 Source: Published/submitted/approved results March 30, 2009. See backup for additional details

45

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Crash Simulation Analysis using LS-DYNA* - cluster results



Comparison Details: –

Comparison based on published results at http://www.topcrunch.org/ as of March 30, 2009.



Cluster Result published by SGI on SGI Altix ICE 8200EX* server based cluster



Benchmark details: –







LS-DYNA is a general purpose transient dynamic finite element program capable of simulating complex real world problems, for use in various industries, including Automobile Design Design, Aerospace Aerospace, Manufacturing, and Bioengineering. The chart for 3 Vehicle collision shows cluster benchmark comparisons of automotive crash simulation for 3 vehicle rear end crash. The simulation involves a minivan crashing into rear of a compact car which in turn crashes into rear of a car, mid-size car. The chart for Car2Car shows cluster benchmark comparisons of automotive crash simulation for the head-on crash of two minivans (Car2Car). Results were shown for increasing number of cluster nodes from one to sixteen. sixteen

Relative Performance Higher is better

Relative Performance Higher is better

Excellent cluster scaling g on LS-DYNA workoads Source: Published/submitted/approved results March 30, 2009. See backup for additional details

46

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Computation Fluid Dynamics Performance using ANSYS* FLUENT* - cluster results •

Comparison Details: –

Comparison based on published results at www.fluent.com/software/fluent/fl6bench/ fl6bench_6.4.x/index.htm as of March 30, 2009



Xeon X5570 performance result published by Intel on SuperMicro SuperMicro* X8DTN+ platform



Benchmark details: –

ANSYS FLUENT12 is a commercial computational fluid dynamics (CFD) package and is a product of ANSYS Inc. Inc It has broad physical modeling capabilities well suited to a variety of industrial applications. Fluent provides a set of standardized benchmarks covering a range of fluid flow simulations and software features. Aircraft_2m and sedan 4m are two test case in the Fluent sedan_4m 12.0 benchmark suite. Results were shown for increasing number of cluster nodes from one to eight. Performance metric is called Solver rating

Relative Performance Higher is better

Relative Performance Higher is better

Strong g cluster leadership p on ANSYS FLUENT workloads Source: Published/submitted/approved results March 30, 2009. See backup for additional details

47

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms WRF v3.0.1 performance on a multi-node cluster



Key Details Cluster configuration, Comparison on equal number of Nodes, Each node contains two quad-core cpus. 1 thread/core Result provided by SGI on SGI Altix ICE 8200EX* server based cluster as of March 30, 2009



Benchmark notes •

WRF – (Weather Research and Forecasting) Model from a group of organizations headed by National Center for Atmospheric Research (NCAR) – a next-generation mesocale numerical weather prediction system designed to serve both operational forecasting and atmospheric research needs.

Excellent cluster scaling g on WRF Workload Source: Published/submitted/approved results March 30, 2009. See backup for additional details

48

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms MM5 v4.7.4 performance on a multi-node cluster

Relative Performance Higher is better



Key Details Cluster configuration, Comparison on equal number of Nodes, Each node contains two quad-core cpus. 1 thread/core Result provided by SGI on SGI Altix ICE 8200EX* server based cluster as of March 30, 2009



Benchmark notes The MM5 (short Th ( h t for f Fifth-Generation Fifth G ti NCAR/P NCAR/Penn St State t Mesoscale M l Model) M d l) is i a regional i l mesoscale l model d l used d for f creating ti weather forecasts and climate projections. It is maintained by Penn State University and the National Center for Atmospheric Research.

Excellent cluster scaling g on MM5 Workload Source: Published/submitted/approved results March 30, 2009. See backup for additional details

49

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

50

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms ISV Application Performance

ISV Ansys* ERDAS*

Segment HPC Enterprise

Market Application

Xeon 5500 vs. 5400 series

CFX11* is the leading software used for computational fluid dynamics (CFD) simulations in manufacturing segment

+88%

ERDAS ERMapper* Image Web Server 2009 application is one of the fastest and flexible image server in the market

+65%

ESI Group*

HPC

PAM CRASH* iis a test and PAM-CRASH* d occupant safety f simulation i l i software

+50%

Exit Games*

Telco

Neutron® is a managed service platform for mobile multiplayer and online games based on .NET*

+80%

Epic* p

Healthcare

Epic server farm acts as a presentation layer in the deployment of Epic electronic medical record (EMR) solution

+82%

Giant*

Enterprise

Giant is the one of the biggest online game vendor in China. Juren* online game is the newest game developed by Giant

+160%

IBM*

Enterprise

IBM® DB2® 9.5 Universal Database: The industry’s leading hybrid relational and XML data server

+60%

IBM*

Enterprise

IBM® DB2® 9.5 Universal Database: The industry’s leading hybrid relational and XML data server under virtualization

+56%

IBM*

Enterprise

IBM® Informix® Dynamic Server 11.5: Blazing OLTP performance, reliability and nearly hands-free administration

+84%

IBM* IBM

Enterprise

IBM® solidDB® 6.3: In-memory relational database that d li delivers up tto 10 10x performance f over di disk-based kb dd databases t b

+87% 87%

Outstanding Performance Gains on ISV Applications Source: Published/submitted/approved results March 30, 2009. See backup for additional details

51

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms ISV Application Performance

ISV Image Analyzer*

Segment Enterprise

Market Application Image Analyzer is a robust, real-time software application that inspects digital images for illicit content

Xeon 5500 vs. 5400 series

+100%

Infowrap*

HPC

Inforwrap High Performance Cluster (large data set on quantum mechanics) provides power for massive computation

+87%

Infowrap*

HPC

Infowrap High Performance Cluster (small data set on quantum mechanics) provides power for massive computation

+129%

Infowrap*

HPC

Infowrap High Performance Cluster (for weather forecasting) provides power for massive computation

+155%

Intense*

Telco

IECCM* is a cross-platform output management solution that is capable of high-volume high volume high speed processing reports

+114%

Intense*

Telco

IECCM* cross-platform output management solution is capable of high-volume high speed processing under virtualization

+150%

InterSystems*

Enterprise

InterSystems Cache* is a high-performance object database for Electronic Medical Record (EMR)

+63%

Kingdee*

Enterprise

Kingdee APUSIC* is one of the leading J2EE middleware product for large- and medium-sized enterprise

+93%

Kingdee*

Enterprise

Kingdee ERP product EAS* is top brand large- and mediumsized enterprise management software under virtualization

+142%

Kingdom* Kingdom

Enterprise

Kingdom is the leading solution provider of centric stock t transaction ti system t

+141% 141%

Outstanding Performance Gains on ISV Applications Source: Published/submitted/approved results March 30, 2009. See backup for additional details

52

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms ISV Application Performance

ISV

Segment

Market Application

Xeon 5500 vs. 5400 series

Kingsoft*

Enterprise

Kingsoft JXIII Online* Game Server is next generation online game under virtualized environments

+98%

Kingsoft*

Enterprise

Kingsoft JXIII Online* Game Server is next generation online game

+76%

Mediaware*

Enterprise

Vid C Video Conversion i S Server application li i iis the h b backbone kb off Mediaware* InStream* which converts video media formats

+68%

Neowiz*

Enterprise

Pmang is well known internet game portal that services many client and web games using virtualization

+100%

Neusoft*

Healthcare

Neusoft key digital health solution focus is on X-ray computed tomography medical image processing

+131%

Neusoft*

Telco

Neusoft Telecom BSS* is a traditional BSS system for bill management in telecom industry under virtualization

+112%

Neusoft*

Telco

Neusoft Telecom BSS* is a traditional BSS system for bill management in telecom industry under virtualization

+115%

NHN Corp*

Enterprise

NHN Cubrid* targets database platforms for Internet portal service

+295%

Qlick Tech*

Enterprise

QlikView provides business intelligence software using “In memory” technology

+36%

SAS* SAS

Enterprise

SAS Forecast Server* accurately forecasts processes over ti time to t identify id tif ttrends d and d flfluctuations t ti for f forecast f t planning l i

+80% 80%

Outstanding Performance Gains on ISV Applications Source: Published/submitted/approved results March 30, 2009. See backup for additional details

53

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms ISV Application Performance Segment

Market Application

Xeon 5500 vs. 5400 series

SAP*

Enterprise

SAP NetWeaver® Business Intelligence Accelerator speeds BI systems, compressing and distributing data on a blade server

+51%

SAP*

Enterprise

A leading ERP application processing sales and distribution transactions under virtualization

+71%

ISV

ECLIPSE300* compositional ii l simulator i l d describes ib reservoir i flfluid id phase behavior or depth compositional changes

Schlumberger*

HPC

Sungard*

FSI

Supcon* p

p Enterprise

APC-Sensor* intelligent soft sensor and flow-sheet calculation software uses neural networks & first principle modeling

TongTech*

Enterprise

TongTech is No.1 middleware vendor in PRC, and its products include several middleware platforms and application packages

UFIDA*

Enterprise

UFIDA NC* is a J2EE based ERP solution suited for largescale companies

+230%

UFIDA*

Enterprise

UFIDA U9* is a next generation SOA based enterprise management software

+112%

UFIDA*

Enterprise

UFIDA Online software is a SaaS hosting platform targeted for large scale middle and small companies

+237%

Vital Images* Images

Healthcare

Brain Perfusion 4DCT is an imaging application for constructing i images th thatt b becomes available il bl after ft C Cardiac di and dB Brain i S Scans

+77%

BancWare Focus ALM* is a risk management solution for financial institutions

+213% +38% +191% +95%

Outstanding Performance Gains on ISV Applications Source: Published/submitted/approved results March 30, 2009. See backup for additional details

54

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

55

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms 1P to 2P Scaling – Benefits of full configuration

• Compares p p performance of benchmarks on same server platform with one and two processors

Relative Performance Higher is better

• DP capable servers with only one processor installed are not achieving maximum ROI • Platform built with significant headroom to accommodate the second processor • S Second d processor provides id significant i ifi t performance benefit for a small increase to overall system cost • No need to disrupt field for upgrade later if i iti ll configured initially fi d with ith ttwo processors

Source: Published/measured results March 30, 2009. See backup for additional details

Maximize your Server’s potential by adding the 2nd processor Source: Published/submitted/approved results March 30, 2009. See backup for additional details

56

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

57

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms Integer Throughput performance on SPECint*_rate_base2006

Data Source: Published/measured/submitted results as of March 30 30, 2009 2009. See backup for details/ details/. Comparison are for different family of processors of 2P Server platforms - Xeon 5100, 5400, 5500, Opteron 2300 series

Hi h iis b Higher better tt

DC

SC Xeon® 5100 Series Xeon (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Opteron* 2300 Opteron Series (“Barcelona”)

Opteron* 2300 Opteron Series (“Shanghai”)

Even lowest Quad-Core bin of Xeon 5500 series is better than the fastest Opteron! Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

58

Opteron 23xx – Quad-Core AMD Opteron* Model 23xx Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating-Point Throughput performance on SPECfp*_rate_base2006

Data Source: Published/measured/submitted results as of March 30 30, 2009 2009. See backup for details/ details/. Comparison are for different family of processors of 2P Server platforms - Xeon 5100, 5400, 5500, Opteron 2300 series

Hi h iis b Higher better tt

DC

SC Xeon® 5100 Series Xeon (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Opteron* 2300 Opteron Series (“Barcelona”)

Opteron* 2300 Opteron Series (“Shanghai”)

Even lowest Quad-Core bin of Xeon 5500 series is better than the fastest Opteron! Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

59

Opteron 23xx – Quad-Core AMD Opteron* Model 23xx Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer “SPEED” performance on SPECint*_base2006

Hi h iis b Higher better tt

DC

SC Xeon

Xeon® 5100 Series (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Opteron* 2300 Opteron Series (“Shanghai”)

Xeon 5500 series for Single Task Integer Performance Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

60

Opteron 23xx – Quad-Core AMD Opteron* Model 23xx Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point “SPEED” performance on SPECfp*_base2006

Hi h iis b Higher better tt

Data Source: Published/measured/submitted results as of March 30 30, 2009 2009. See backup for details/ details/. Comparison are for different family of processors of 2P Server platforms - Xeon 5100, 5400, 5500, Opteron 2300 series

DC

SC Xeon

Xeon® 5100 Series (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Opteron* 2300 Opteron Series (“Shanghai”)

Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Xeon 5500 series for Single Task Floating Point Performance Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

61

Opteron 23xx – Quad-Core AMD Opteron* Model 23xx Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java performance on SPECjbb*2005

Data Source: Published/measured/submitted results as of March 30 30, 2009 2009. See backup for details/ details/. Comparison are for different family of processors of 2P Server platforms - Xeon 5100, 5400, 5500, Opteron 2300 series

Higher Hi h iis b better tt Business Operations Per Second

DC

SC Xeon® 5100 Series Xeon (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Opteron* 2300 Opteron Series (“Barcelona”)

Opteron* 2300 Opteron Series (“Shanghai”)

Xeon 5500 series for Java Servers Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

62

Opteron 23xx – Quad-Core AMD Opteron* Model 23xx Performance of skus with different TDPs of Xeon 5400 series is estimated to be same. Eg. L5430 = E5430; X5450 = E5450

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficienct performance on SPECpower*_ssj 2008

Data Source: Intel measured results as of March 30, 2009 See backup for details 2009.

Relative R l ti P Performance f Higher is better

DC

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem (“Nehalem--EP”)

Xeon 5500 series for Energy Efficient Servers Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”)

63

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Dense Floating-Point performance on LINPACK

Higher Hi h iis b better tt Gflops

Data Source: Measured results as of March 30, 2009.

DC

SC Xeon® 5100 Series Xeon (“Woodcrest”)

Xeon® 5400 Series (“Harpertown (“ Harpertown”) ”)

Xeon® 5500 Series (“Nehalem--EP”) (“Nehalem

Xeon 5500 series for Scientific Computing Xeon 51xx – Dual-Core Intel® Xeon® Processor 51xx (“Woodcrest”) Performance of skus with different TDPs of Xeon 5400 series is Xeon 54xx – Quad-Core Intel® Xeon® Processor 54xx (“Harpertown”) estimated to be same. Eg. L5430 = E5430; X5450 = E5450 Xeon 55xx – Intel® Xeon® Processor 55xx (“Nehalem-EP”) Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 64 software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

65

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms Integer Throughput performance on SPECint*_rate_base2006

Relative Performance Higher is better

Xeon® 5160

Xeon® 5365

Xeon® 5470

Xeon® 5570

Consolidate your IT infrastructure with Xeon 5500 series Server performance based on best publically available SPECint_rate_base2006* results from www.spec.org as of March 30, 2009. For the result s from 20012004 (first five bars in the chart) performance scores are based on Intel internal estimates.

66

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point Throughput performance on SPECfp*_rate_base2006

Relative Performance Higher is better

Xeon® 5160

Xeon® 5365

Xeon® 5482

Xeon® 5570

Consolidate your IT infrastructure with Xeon 5500 series Server performance based on best publically available SPECfp_rate_base2006* results from www.spec.org as of March 30, 2009. For the result s from 20012004 (first five bars in the chart) performance scores are based on Intel internal estimates.

67

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Database performance on TPC*-C

Higher is better

Measured in Transactions Per Minute (tpmC)

Xeon® 5160

Xeon® 5365

Xeon® 5460

Xeon® 5570

Consolidate your IT infrastructure with Xeon 5500 series Source: Published/submitted/approved results March 30, 2009. See backup for additional details

68

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Enterprise Resource Planning performance on SAP-SD* 2-Tier

Higher is better

Measured in Number of SD users

Xeon® 5160

Xeon® 5365

Xeon® 5470

Xeon® 5570

Consolidate your IT infrastructure with Xeon 5500 series Source: Published/submitted/approved results March 30, 2009. See backup for additional details

69

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms eCommerce performance on SPECjbb*2005

Higher is better

Business Operations Per Second

Xeon® 5160

Xeon® 5365

Xeon® 5470

Xeon® 5570

Consolidate your IT infrastructure with Xeon 5500 series Server performance based on best publically available SPECijbb2005* results from www.spec.org as of March 30, 2009.

70

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Table of Contents

71

What’s new

Platform Overview, SKU Lineup

Summary

Key Messages, Performance highlights

Server Performance

Comparison to prior generation Xeon platforms and competition on Industry standard benchmarks

Energy Efficiency

Performance Per Watt Comparison

Virtualization

Performance comparison on different VMMs & benchmarks

HPC Performance

Comparison against prior generation Xeon platforms and vs competition on Single Node and Cluster HPC applications

Application Performance

Performance comparison on “Real-World” ISV applications

Scalability

Server Platform Scalability (1P Æ 2P Scaling)

Down Bin Performance

Comparison on all bins on current and prior generation platforms on select benchmarks

Server Consolidation

Comparison across older generation platforms to demonstrate excellent opportunity for Server Consolidation

Other

Memory bandwidth for various configuration

Intel® Xeon® Processor 5500 series based Server platforms Stream Bandwidth for Xeon X5570 processor

StMaximum B d B/W: idth – Mbytes/Sec Mb t /S (Triad) (T i d) • Stream M i Bandwidth B/W – DDR3 1333 across 3 channels 36588 Higher –is better Up to 1 DPC (6 DIMMs total) 33203 – Max capacity: 48 GB

10.6 GB/s 10.6

1333 MHz memory

CPU

CPU

10.6

CPUs X5550 and above

+274% 1066 MHz memory

27208

• General purpose:

8.5 GB/s

– DDR3 1066 across 3 channels – Up to 2 DPC (12 DIMMs total) – Max capacity: 96GB

8.5

9776

64 6.4

(DPC – Dimms Per Channel)

CPU

CPU

6.4

– Up to 3 DPC (18 DIMMs total) – Max capacity: 144GB NHM 2.93/ 1333 MHz mem/1 DPC

E5520 and above

800 MHz memory

6.4 GB/s

6102 – DDR3 800 across 3 channels

HTN 3.00/ NHM 2.93/ 800 NHM 2.93/ SB1600/ 800 MHz mem/3 1066 MHz MHz mem DPC mem/2 DPC

CPU

85 8.5

• Maximum capacity:

HTN 3.16/ BF1333/ 667 MHz mem

CPU

All SKUs

Nehalem-EP memory Bandwidth for different configuration Memory speed 800 MHz 1066 MHz 1333 MHz 1 DPC 2 DPC 3 DPC 1 DPC 2 DPC 1 DPC Stream Triad 27748 26565 27208 33723 33203 36588

Massive Increase in Platform Bandwidth Source: Intel internal measurement – March 2009

72

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Matching Processor to Memory: Selection Guidance

Memory Requirement

Max Bandwidth

Balanced Performance

Max Capacity

HPC

General Purpose Enterprise workloads

Virtualized Environment

Example Usages Memory Technology

DDR3 1333

DDR3 1066

DDR3 800

48 GB max mem Max of 6 Dimms

96 GB max mem Max of 12 Dimms

144 GB max mem Max of 18 Dimms

X5570

X5570

Available CPU Options

Advanced ƒ ƒ ƒ ƒ

95W

8M cache 6.4 GT/s QPI HT Turbo

2 93 GH 2.93 GHz

X5560 1 DPC – 36588 2.80 GHz X5550 2.66 GHz

St d d Standard ƒ ƒ ƒ ƒ

80W

8M cache 5.86 GT/s QPI HT Turbo

2 93 GHz 2.93

2 DPC – 33203

X5560 2.80 GHz

X5560 2 DPC – 26565 2.80 GHz

X5550

X5550 1 DPC – 27748

1 DPC – 33723 2.66 GHz

ƒ 4M cache ƒ 4.8 GT/s QPI

80W

All data shown are Stream-Triad bandwidth in MB/Sec Platform is two socket with a 3 Channels Per Socket DPC – Dimm(s) Di ( ) Per P Channel Ch l

2.66 GHz

2.53 GHz

E5540 3 DPC –GHz 26305 2.53

E5530

E5530 2 DPC – 25884

E5540 2 DPC – 30912 2.40 GHz

1 DPC – 31218

E5520 2.26 GHz

Basic

X5570

3 DPC –GH 27208 2 93 2.93 GHz

2.40 GHz

E5520 1 DPC – 26750 2.26 GHz

E5506

3 DPC –GHz 24052 2.13

E5504 2 DPC – 23866 2.00 GHz E5502 1 DPC – 24265 1.86 GHz (2C)

Source: Intel internal measurement – March 2009

73

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Other Memory variations – Eight Dimm configuration

• Measured for DDR3-1066 and DDR3-800 memory • A balanced configuration needs to be a multiple of six • Any system configuration with eight Dimms is an unbalanced configuration (assuming 3 channels per cpu) • Data provided below since some tenders require this – This is not the recommended configuration for Xeon 5500 based platforms • Measured with Xeon E5540 (NHM-EP 2.53/5.86QPI) • Two variations shows – 2 2-2-0 2 0 and 2 2-1-1 1 1

Triad Total Dimms Dimm/Socket Socket 0/Ch 0  Socket 0/Ch 1 Socket 0/Ch 2 Socket 1/Ch 0  Socket 1/Ch 1 Socket 1/Ch 2

2‐2‐0/2‐2‐0 19510.2

DDR3‐800 2‐1‐1/2‐1‐1 19961.4

2‐2‐2/2‐2‐2 25883.9

8 4 2 Dimms 2 Dimms 0 Dimms 2 Dimms 2 Dimms 0 Dimms

8 4 2 Dimms 1 Dimm 1 Dimm 2 Dimms 1 Dimm 1 Dimm

12 6 2 Dimms 2 Dimms 2 Dimms 2 Dimms 2 Dimms 2 Dimms

2‐2‐0/2‐2‐0 25343.4

DDR3‐1066 2‐1‐1/2‐1‐1 25679.0

2‐2‐2/2‐2‐2 30912.4

8 4 2 Dimms 2 Dimms 0 Dimms 2 Dimms 2 Dimms 0 Dimms

8 4 2 Dimms 1 Dimm 1 Dimm 2 Dimms 1 Dimm 1 Dimm

12 6 2 Dimms 2 Dimms 2 Dimms 2 Dimms 2 Dimms 2 Dimms

All data shown are Stream-Triad bandwidth in MB/Sec Source: Intel internal measurement – March 2009

74

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Summary

Performance Energy Efficiency

Virtualization

• Record breaking result on major Server & HPC benchmarks • Up to 3.5x the bandwidth for technical computing† • Up to 2.25x performance for enterprise computing†

• All time high SPECpower*_ssj2008 score • Lower system idle power reduces IT costs • Dynamically turns cores on/off to meet performance needs

• Highest 2S VMmark* score • Exceptional consolidation ratio using Virtualization

Refresh with Intel Xeon 5500 series: Boost Performance, Lower IT Costs, 8-month ROI ®



75

®

Source: Intel internal measurements, January 2009. For notes and disclaimers, see performance and legal information slides at end of this presentation.

Backup

76

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” All comparisons based on published/submitted/approved results as of June 9,2009

SPECint_rate_base2006: Intel® Xeon® processor X5470 based platform details Fujitsu Siemens PRIMERGY* RX200 S4 server platform with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (8x2GB DDR2 PC2-5300F, 2 rank, CAS 5-5-5, with ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 11.0 build 20080730. Referenced as published at 140. For more information see http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080901-05156.html. Intel® Xeon® processor X5570 based platform details ASUS Z8PE-D12X based server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4GB PC3-10600R, CL9-9-9, ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20090131. Referenced as published at 241. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090316-06703.html Quad-Core AMD Opteron* processor model 2393 based platform details Dell Inc. PowerEdge SC1435 (AMD Opteron 2393 SE, 3.10 GHz) 148 124 8 2 4 with two Quad-Core AMD Opteron processors 2393SE 3.10GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dualrank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 124 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07450.html

SPECfp_rate_base2006 Intel® Xeon® processor X5482 based platform details Hewlett-Packard ProLiant DL160 G5p server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz CL5 FB-DIMM), 64-Bit SUSE Linux Enterprise Server 10 SP1, Intel C++ Compiler for Linux32 and Linux64 version 10.1 build 20080730. Referenced as published at 86.4. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05587.html. Intel® Xeon® processor X5570 based platform details http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090406-06932.html Cisco B-200 M1* server platform with two Quad-Core Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB PC3-10600R, 2 rank, CL9-9-9, ECC), SUSE Linux Enterprise Server 11_RC4 x86_64 Kernel 2.6.27.15-2-default, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Results measured at 194 as of March 30, 2009. ASUS Z8PE-D12X based server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 72 GB (18 X 4 GB PC3-10600R, CL=9, SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. References at 197 For more information see: http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07354.html Quad-Core Quad Core AMD Opteron* processor model 2393based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two Quad-Core AMD Opteron processors AMD Opteron 2393 SE, 3.1 GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 112 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090406-06932.html

SPECint_base2006: Intel® Xeon® processor X5470 based platform details Dell PowerEdge 2950 III server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (4x4GB DDR2 667MHz FB Dimm), SUSE Linux Enterprise Server 10* SP1 x86_64, kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see htt // http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05450.html. / 2006/ lt / 2008 4/ 2006 20080929 05450 ht l Intel® Xeon® processor X5570 based platform details ASUS RS700-E6 server system with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Referenced as published at 32.1 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090330-06836.html Quad-Core AMD Opteron* processor 2384 based platform details Dell Inc. PowerEdge M805 with two AMD Opteron2393 SE, 3.10 GHz , 6MB L3 cache, 16 GB (4x4 GB, PC2-6400P CL5) memory, SuSE Linux Enterprise Server 10 (x86_64) SP1, Kernel 2.6.16.46-0.12-smp, PGI Server Complete Version 7.2*, PathScale Compiler Suite Version 3.2*. Referenced as published at 19.1For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-2009042807285 html 07285.html

Slide 1 of 6 77

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” •

All comparisons based on published/submitted/approved results as of June 9 2009



SPECfp_base2006

• •

Intel® Xeon® processor X5482 based platform details IBM System x 3450* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory, 64-Bit SUSE Linux Enterprise Server 10*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081215-06283.html. Intel® Xeon® processor X5570 based platform details Hitachi BladeSymphony BS2000 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR3-1333MHz), RHEL 5.3 x86_64 Kernel 2.6.18-128.el5, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Referenced as published at 39.3. For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-2009031706805.html Quad-Core AMD Opteron* processor model 2389 based platform details HP ProLiant DL385 G5p* platform with two AMD Opteron 2389, 2.9 GHz 6MB L3 cache, 32 GB (8x4 GB, PC2-6400P CL5) memory, Red Hat Enterprise Linux Server release 5.3, Kernel 2.6.18-128.el5 , PGI Server Complete Version 8.0, PathScale Compiler Suite Version 3.2*. Referenced as published at 22.8. For more information see:

• •

• •

http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090427-07257.html



SAP-SD 2-Tier

• •

Intel® Xeon® processor X5470 based platform details HP ProLiant BL460C server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005, SAP ECC Release 6.0 . Referenced as published at 2,518 SD users. Certification number 2008048. Intel® Xeon® processor X5570 based platform details IBM System x3650 M2 Server platform with two Intel Xeon processors X5570, 2.93GHz 8MB L3 cache, 6.4QPI, 48GB memory, Microsoft Windows Server 2003 Enterprise Edition, DB2 9.5, SAP ECC Release 6.0 (2005). Referenced as published at 5,100 SD users. Certification number 2008079 HP Proliant DL385 G5p Server platform* with two Quad-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005 database, SAP ECC Release 6.0. Referenced as published at 2,752 SD Users. Certification number 2008065.

• • •



TPC-C

• •

Intel® Xeon® processor X5460 based platform details HP ProLiant ML370 G5 platform with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Microsoft SQL Server 2005 x64 Enterprise Edition SP2, Microsoft Windows Server 2003 Enterprise x64 Ent. R2. Referenced as published at 275,149 tpmC and $1.44/tpmC; availability date January 7, 2008. For more information see http://tpc.org/tpcc/results/tpcc_result_detail.asp?id=108010701. Intel® Xeon® processor X5570 based platform details HP ProLiant DL370 G6* platform with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 144 GB memory (18x8 GB DDR3). Oracle Database 11g Standard Edition One * with Oracle Enterprise Linux OS*. Referenced as published at 631,766 tpmC and $1.08/tpmC; availability date March 30, 2009. For more information see http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=109033001

• •



TPC-E

• •

Intel® Xeon® processor X5460 based platform details Fujitsu-Siemens PRIMERGY TX300 S4 server platform* with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Microsoft SQL Server 200 x64 Enterprise Edition*, Microsoft Windows Server 2008 Enterprise x64*. Referenced as published at 317.45 tpsE and $523.49/tpsE; availability date August 30, 2008. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=12. Intel® Xeon® processor X5570 based platform details Fujitsu-Siemens PRIMERGY RX300 S5 server platform* with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066) DDR3 1066), Microsoft SQL Server 2008 x64 Enterprise Edition Edition, Microsoft Windows Server 2008 Enterprise x64 x64. Referenced as published at 800tpsE and $343 $343.91/tpsE. 91/tpsE Availability date April 1 1, 2009 2009. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=25

• •

Slide 2 of 6 78

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” •

All comparisons based on published/submitted/approved results as of June 1, 2009



SPECjbb2005

• •

Intel® Xeon® processor X5470 based platform details Fujitsu Siemens PRIMERGY RX200 S4 server platform* with two Intel Xeon processors 5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory, Microsoft Windows Server 2008 Enterprise x64 Edition*, Oracle JRockit 6 P28.0.0 (build P28.0.0-8-109238-1.6.0_05-20090130-1408-windows-x86_64) 4 JVM instances. Referenced as published at 368,034 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090220-00583.html. Intel® Xeon® processor X5570 based platform details IBM Bladecenter HS22 Server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Microsoft Windows Server 2008 Enterprise x64 Edition, IBM J9* 2.4 JRE 1.6.0 (build pwa6460sr5-20090323_04(SR5)) run with 4 JVM instances. Result submitted to www.spec.org for review at 604417 BOPS as of March 30, 2009. Quad-Core AMD Opteron* processor 2393SE based platform details SuperMicro A+ Server 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2393SE 3.1 GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Microsoft Windows Server 2008 Enterprise with SP1 (64-bit), IBM J9 JVM build 2.4, J2RE 1.6.0 IBM J9 2.4. 2 JVM instances. Referenced as published at 408202 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q2/jbb2005-20090406-00704.html

• • • •



SPECweb2005

• •

Intel® Xeon® processor X5460 based platform details HP Proliant DL380 G5 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 32GB memory (8x4G 667MHz ECC DDR2 FB-DIMM), RedHat Enterprise Linux 5 (2.6.18-53.el5), Rock Web Server v1.4.6 x86_64. Referenced as published at 29591. For more information see http://www.spec.org/web2005/results/res2008q1/web2005-20080225-00104.html. Intel® Xeon® processor X5570 based platform details platform* with two Intel Xeon p processors X5570 2.93GHz,, 8MB L3 cache,, 6.4GT/s QPI, Q , 144 GB memoryy (18x8 ( GB DDR3), ), Red Hat Enterprise p Linux 5.2 .Rock Web Server v1.4.7 HP ProLiant DL380 G6 p (x86_64). Result published at 75023. For more information see http://www.spec.org/osg/web2005/results/res2009q2/web2005-20090423-00135.html Quad-Core AMD Opteron* processor model 2384 based platform details HP Proliant DL165 G5 Server platform with two Dual-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, memory 32GB (8x4GB, PC2-6400P CL5), RedHat* Enterprise Linux .5.2 (2.6.1892.el5), Rock Web Server v1.4.7 x86_64. Referenced as published at 39530. For more information see http://www.spec.org/osg/web2005/results/res2009q1/web2005-20090122-00125.html.

• • • •



SPECjAppServer2004

• •

Baseline Intel® Xeon® processor X5460 based platform details: HP Proliant BL460c G1 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 16GB memory (8x2G 667MHz ECC DDR2 FB-DIMM), Oracle Application Server 10G Release 10.1.3.3 - Java Edition, BEA JRockit(R) 6.0 JDK (R27.3.0-106) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.6. Referenced as published at 2056. For more information see http://www.spec.org/osg/jAppServer2004/results/res2007q4/jAppServer2004-20071023-00088.html . Intel® Xeon® processor X5570 based platform details: Dell PowerEdge R610 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (12x2 GB DDR3), Oracle WebLogic Server Standard Edition Release 10.3, Oracle JRockit(R) 6.0 JDK (R27.6.0-50) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.7, Result submitted to www.spec.org for review at 3975 as of March 30, 2009.

• •



SPECjvm2008:

• •

Intel® Xeon® processor X5470 based platform details: Supermicro X7DBE +* server platform with two Intel® Xeon® processors X5470 3.333GHz, 12MB L2 Cache, 1333MHz FSB, 16GB memory (8x2GB 667MHz DDR2 FB-DIMM), Red Hat Enterprise Linux 5.3 (kernel: 2.6.18-128.el5), JDK 1.6.0_06-p-b04. Intel internal measurement. Measured at 179.53 SPECjvm2008 base ops/m as of March 30, 2009 Intel® Xeon® processor X5570 based platform details: Sun Blade X6270* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR3-1333MHz), OpenSolaris* 2008.11, ZFS, JDK 1.6.0_14. Result submitted to www.spec.org for review at 317.13 SPECjvm2008 base ops/m as of April 14, 2009.

• •

Slide 3 of 6 79

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” • • • •

• • • •

• • • • • • • • •

All comparisons based on published/submitted/approved results as of June 1, 2009 SPECpower_ssj2008 Intel® Xeon® processor L5430 based platform details Powerleader PR2510D2 server platform* with two Intel Xeon processors L5430 2.33GHz, 12MB L2 cache, 1333MHz FSB, 8GB memory, Oracle JRockit* (build P27.5.0-5_o_CR371811_CR374296100684 1 6 0 03 20080702 1651 i d 100684-1.6.0_03-20080702-1651-windows-x86_64, 86 64 compiled il d mode). d ) Published P bli h d att 1135 ssj_ops/watt. j / tt F For more iinformation f ti see: http://www.spec.org/power_ssj2008/results/res2008q4/power_ssj2008htt // / j2008/ lt / 2008 4/ j2008 20081007-00086.html. Intel® Xeon® processor X5570 based platform details IBM System x3650 M2* server platform with two Intel Xeon processor X5570, 2.93GHz, 8 GB (4 x 2) memory, Microsoft Windows Server 2008 Enterprise* OS. IBM J9 Java™6 Runtime Environment JVM. Result submitted to www.spec.org at 1977 ssj_ops/watt for more information see: http://www.spec.org/power_ssj2008/results/res2009q2/power_ssj2008-20090519-00165.html Quad-Core AMD Opteron* processor model 2376HE based platform based details Supermicro Inc. 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2376HE 2.30GHz, 6MB L3 cache, 16GB memory, IBM J9 VM (build 2.4, J2RE 1.6.0 IBM J9 2.4 Linux amd64-64 jvmxa6460-20081105_25433 (JIT enabled, AOT enabled), Published at 1044 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2009q1/power_ssj200820090210-00111 20090210 00111.html. html VMmark: Intel® Xeon® processor X5470 based platform details HP Proliant* ML370 G5 server platform with two Intel Xeon processors X5470 3.33GHz, 2x6MB L2 cache, 1333MHz FSB, 48GB memory, VMware ESX V3.5. Update 3 Published at 9.15@7 tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-10-09-ML370.pdf Intel® Xeon® processor X5570 based platform details HP ProLiant BL490 G6 VMware ESX v4.0 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 24.24@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-HP-2009-05-19-BL490G6.pdf HP ProLiant ProLiant* DL385 G5p platform with two AMD Opteron 2384 2384, 6MB L3 Cache Cache, 64GB memory memory. VMware ESX V3 V3.5. 5 Update 3 Published at 11.28@8 11 28@8 tiles tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-12-30-DL385.pdf. SPECompM2001 – SPECompM results on Intel® Xeon® processor 5500 series-based server were obtained using 16 OpenMP threads over two sockets, on Intel® Xeon® processor 5400 series and AMD Opteron 2300 series-based server 8 OpenMP threads over two sockets were used. Baseline Intel® Xeon® processor E5472 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory (8x4GB 800MHz DDR2 FB-DIMM), SUSE LINUX 10.1* (X86-64) (Linux 2.6.16.13-4-smp). Binaries built with Intel Compiler 10.1. Referenced as published at 17187. (SPECompMbase2001). For more information see http://www.spec.org/omp/results/res2007q4/omp2001-20071107-00274.html.

• •

Intel® Xeon® processor X5570 based platform details: Sun Blade X6275 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), OpenMP and Automatic parallel Operating System: OpenSolaris 2008.11 Compiler: Sun Studio 12 Update 1 (backend build 20090309) Result published at 43777 (SPECompMbase2001) for more information see: http://www.spec.org/omp/results/res2009q2/omp2001-20090413-00350.html



Quad-Core AMD Opteron processor 2384 based platform* details: Supermicro H8DMU Server platform* with two Quad-Core AMD Opteron processors 2386SE* 2.80GHz, 6MB L3 cache, 16GB memory (8x2GB, PC2-6400, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit, Binaries built with PathScale Compiler Suite*, Release 3.1. Referenced as published at 22678 (SPECompMbase2001). For more information http://www.spec.org/omp/results/res2008q4/omp2001-20081021-00320.html.



SPEComp*Lbase2001 - SPECompL results on Intel® Xeon® processor 5500 series-based server were obtained using 16 OpenMP threads over two sockets, on Intel® Xeon® processor 5400 seriesbased server 8 OpenMP threads over two sockets were used. Baseline Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory (8x4GB 800MHz DDR2 FB-DIMM), SUSE LINUX 10 SP2. Binaries built with Intel Compiler 11.0. Result submitted to www.spec.org for review at 81332 (SPECompLbase2001) as of March 30, 2009. Intel® Xeon® processor X5570 based platform details: Sun Blade X6275 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), OpenMP and Automatic parallel Operating System: OpenSolaris 2008 2008.11 11 Compiler: Sun Studio 12 Update 1 (backend build 20090309) Result published at 237781 (SPECompLbase2001) for more information see: http://www.spec.org/omp/results/res2009q2/omp2001-20090413-00349.html .

• • • •



Slide 4 of 6 80

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” •

All comparisons based on published/submitted/approved results as of March 30, 2009



Finite Element analysis using ANSYS FLUENT*:



Comparison based on published/submitted results to www.fluent.com/software/fluent/fl6bench/fl6bench_6.4.x/index.htm as of March 30, 2009. All comparisons were using g results run on 8 cores within a single machine on dual socket quad-core servers. Baseline Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.13)*. Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchmarks. Intel® Xeon® processor X5570 based platform details: SGI Altix ICE 8200EX* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.9) Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchmarks. Quad-Core AMD Opteron* processor model 2384 platform based details:Server platform with two AMD Opteron 2384 processor 2.7GHz, 6MB L3 cache, Linux OS. Performance measured using Fluent Version 12 12.0 0 Beta Beta. (Version 12 12.0.7) 0 7) Six individual benchmarks are shown as a measure of single node performance performance. "Overall" Overall performance is the geometric mean of the six individual benchmarks. benchmarks









Crash Simulation analysis using LS-DYNA* (Single Node):



Comparison based on published/submitted results to http://www.topcrunch.org/ as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux* 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car.







Multiphysics Finite Element Analysis using ANSYS* -

• •

Comparison based on published/submitted results to www.ansys.com/services/hardware-support-db.htm as of March 30, 2009. Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), FB DIMM), 64-bit 64 bit RedHat Enterprise Linux 5.1 5.1*.. Performance measured using ANSYS* ANSYS Mechanical Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared* Ansys Shared consists of a suite of 8 workloads and Ansys-Distributed* consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using ANSYS* Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared consists of a suite of 8 workloads and AnsysDistributed consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison.





Unified Finite Element Analysis using Abaqus*:

• •

Comparison based on published/submitted results to www.simulia.com/support/v68/v68_performance.php as of March 30, 2009. Geo mean of S2A, S4B, S4D, S6 used for comparison I t l® Xeon® Intel® X ® processor X5482 based b d platform l tf details: d t il Supermicro S i X7DB8+* X7DB8 * server platform l tf with ith ttwo IIntel® t l® X Xeon® ® processors X5482 3 3.20GHz, 20GH 12MB L2 cache, h 1600MHz 1600MH FSB, FSB 16GB memory (8x2GB (8 2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Abaqus/Standard* and Abaqus/Explicit v6.8*-EF1. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using Abaqus/Standard and Abaqus/Explicit v6.8-EF1. Quad-Core AMD Opteron* processor 2384 based platform details: HP ProLiant* BL465 G5 platform with two AMD Opteron 2384, 6MB L3 cache, 16 GB memory, SuSE Linux Enterprise Server 10 (x86_64).

• •

Slide 5 of 6 81

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Server & HPC Performance comparison to Xeon 5400 series AMD Opteron 2300 Series “Shanghai” •

All comparisons based on published/submitted/approved results as of March 30, 2009



Computational Fluid Dynamics analysis using Star-CD* (Single Node) :



Comparison based on published/submitted results to http://www.cd-adapco.com/products/STAR-CD/performance/406/index.html as of March 30, 2009. All comparisons were using g results run on 8 cores within a single machine on dual socket quad-core servers. Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using STAR-CD v4.06. Same configuration used for all both benchmark results - A-Class and C-Class. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI* 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using STAR-CD v4.06. Same configuration used for all both benchmark results - A-Class and C-Class.

• •



Reservoir Simulation using Schlumberger Eclipse*



Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. (64-bit) and Eclipse version 2008.1 software. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 24GB (12x2GB 1066MHz DDR3) memory, 64-bit RedHat Enterprise Linux 5.3. Eclipse version 2008.1 software.





Reservoir Simulation using Landmark Nexus*



Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB ), 64-bit RedHat Enterprise p Linux 5.2*. ((64-bit)) . Landmark Nexus R5000 software*. 800MHz DDR2 FB-DIMM), Intel® Xeon® processor X5560 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5560 2.80GHz, 8MB L3 cache, QPI 6.4 MT/sec, 12GB memory, 64-bit RedHat Enterprise Linux 5.3. Landmark Nexus R5000 software.





Reservoir Simulation using CMG* IMEX*



Intel® Xeon® processor X5482 based platform details: Dell Precision T7400 platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB RedHat Enterprise Linux 5.2*. (64-bit) CMG IMEX, Version 2008.11. Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 18GB memory. RedHat E t Enterprise i Li Linux 5 5.3. 3 (64 (64-bit) bit) CMG IMEX, IMEX Version V i 2008.11. 2008 11





MM5 v4.7.4 - t3a and WRF v3.0.1 - 12km CONUS :



Comparison based on measured results as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Same platform used for both benchmarks Baseline Intel® Xeon® processor X5482 based platform details: SGI Altix ICE 8200EX* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*. te ® Xeon® eo ® p processor ocesso X5570 55 0 based p platform at o deta details: s SG SGI Altix t ICE C 8200EX* 8 00 server se e platform p at o with t two t o Intel te Xeon eo processors p ocesso s X5570 55 0 2.93GHz, 93G , 8MB 8 L3 3 cache, cac e, QPI Q 6.4 6 MT/sec, /sec, 24GB G memory e o y ((12x2GB G Intel® 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*..

• •

Slide 6 of 6 82

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Server Performance comparison to Six-Core Istanbul - Backup

Current at of June 11.,2009 SPECint_rate2006 Xeon X5570 based platform details • ASUS Z8PE-D12X based server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4GB PC3-10600R, CL9-9-9, ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20090131. Referenced as published at 241. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090316-06703.htm Six-Core Opteron 2434 based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two six-core AMD Opteron processors 2434 2.6GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Red Hat Enterprise Linux Server release 5.3, PGI Server Complete Version 8.0 x86 Open64 4.2.2 Compiler Suite (from AMD) publishes at 161 http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07357.html SPECfp_rate2006 Intel® Xeon® processor X5570 based platform details ASUS Z8PE-D12X Z8PE D12X based server platform with two Intel Xeon processors X5570 2 2.93GHz, 93GHz 8MB L3 cache cache, 6 6.4GT/s 4GT/s QPI QPI, 72 GB (18 X 4 GB PC3 PC3-10600R, 10600R CL=9 CL=9, SUSE Linux Enterprise Server 10 SP2 x86 x86_64 64 Kernel 2.6.16.60-0.21-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. References at 197 For more information see: http://www.spec.org/cpu2006/results/res2009q2/cpu2006-2009051107354.html Alternative Intel® Xeon® processor X5570 based platform result Cisco B-200 M1* server platform with two Quad-Core Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB PC3-10600R, 2 rank, CL9-9-9, ECC), SUSE Linux Enterprise Server 11_RC4 x86_64 Kernel 2.6.27.15-2-default, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Results measured at 194 as of March 30, 2009. Six-Core Opteron 2434 based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two six-core AMD Opteron processors 2434 2.6GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Red Hat Enterprise Linux Server release 5.3, PGI Server Complete Version 8.0 x86 Open64 4.2.2 Compiler Suite (from AMD) publishes at 129 for more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07358.html SPECjbb2005 Intel® Xeon® processor X5570 based platform details IBM Bladecenter HS22 Server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Microsoft Windows Server 2008 Enterprise x64 Edition, IBM J9* 2.4 JRE 1.6.0 (build pwa6460sr5-20090323_04(SR5)) run with 4 JVM instances. Result submitted to www.spec.org for review at 604417 BOPS as of March 30, 2009. Six-Core AMD Opteron* processor 2434 based platform details

Hewlett-Packard Company, ProLiant DL165 G6 with two six-core AMD Opteron 2435, 2.6 GHz, 32768 MB, 8 x 4 GB PC2-6400P CL5, 64 KB I + 64 KB D on chip, per core, Secondary cache 512 KB I+D on chip, per core, 6 MB I+D on chip, per chip, JVM Version Oracle JRockit(R) 6 P28.0.0 (build P28.0.0-29-114096-1.6.0_11-20090427-1759-windows-x86_64, compiled mode) published at 484 484,297, 297 http://www.spec.org/osg/jbb2005/results/res2009q2/jbb2005-20090519-00728.html http://www spec org/osg/jbb2005/results/res2009q2/jbb2005 20090519 00728 html SAP SD 2-Tier Intel® Xeon® processor X5570 based platform details Sun Fire X4270 Server with two Intel Xeon processors X5570, 2.93GHz 8MB L3 cache, 6.4QPI, 48GB memory, Solaris 10, Oracle 10g, SAP ECC Release 6.0 Enhancement Pack 4 (Unicode) . Referenced as published at 3,700 SD users. Certification number 2009005. Six-Core AMD Opteron* processor model 2434 based platform details HP Proliant DL385 G5p Server platform* with two Six-Core AMD Opteron processors 2434, 2.60GHz, 6MB L3 cache, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005 database, SAP ECC Release 6.0 Enhancement Pack 4 (Unicode) , Referenced as published at 2355 SD Users. Certification number 2009020 VMmark* AMD Opteron* processor model 2384 based platform details HP ProLiant* DL385 G5p platform with two AMD Opteron 2434, 6MB L3 Cache, 32GB memory. VMware ESX V4.0 published at 15.54@11 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-HP-2009-06-02-dl385g6.pdf HP HP ProLiant BL490 G6 VMware ESX v4.0 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 24.24@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-HP-2009-05-19-BL490G6.pdf

83

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Server Refresh opportunity

• 8 month ROI claim estimated based on comparison between 2S Single Core Intel® Xeon® 3.80 with 2M L2 Cache and 2S Intel® Xeon® X5570 based servers. Calculation includes analysis based on performance, power, cooling, electricity rates, operating system annual license costs and estimated server costs. This assumes 8kW racks, $0.10 per kWh, cooling costs are 2x the server power consumption costs, operating system license cost of $900/year per server, per server cost of $6900 based on estimated list prices and estimated server utilization rates. All dollar figures are approximate. Performance and power comparisons are based on measured SPECjbb2005* benchmark results (Intel Corporation Feb 2009). Platform power was measured during the steady state window of the benchmark run and at idle. Performance gain compared to baseline was as 9x 9 while hile the platform platfo m power po e was as 0.8x. 08 – Baseline platform: Intel server platform with two 64-bit Intel Xeon Processor 3.80Ghz with 2M L2 Cache, 800 FSB, 8x1GB DDR2-400 memory, 1 hard drive, 1 power supply, Microsoft* Windows* Server 2003 Ent. SP1, BEA* JRockit* build P27.4.0-windows-x86_64 run with 2 JVM instances – New platform: Intel server platform with two quad-core Intel® Xeon® processor X5570, 2.93 GHz, 8 8MB L3 3 cache, h 6.4QPI, 6 Q 24GB 2 G memory (6x2GB (6 2G DDR3-1333), 3 333) 1 hard h d drive, d 1 power supply, l Microsoft Windows Server 2008 Ent. SP1, BEA JRockit build P27.4.0-windows-x86_64 run with 2 JVM instances

• Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations. Limitations

84

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer Throughput performance on SPECint*_rate_base2006



Integer Throughput Performance on SPECint*_rate_base2006 benchmark

• • •

Comparison p based on p published/submitted results to www.spec.org p g as of June 9,, 2009. Intel® Xeon® processor 3.80 with 2M L2 cache based platform details HP* ProLiant* DL380 G4 server platform with two 64-bit Intel Xeon processor 3.80GHz with 2M L2 cache, 800 FSB, 8GB memory, Microsoft Windows Server 2003* Ent. SP1, Intel® C++ Compiler 9.1 for 32-bit apps, Build 20060323Z Package ID: W_CC_P_9.1.020. Referenced as published at 20.9. For more information see http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00027.html. Intel® Xeon® processor 5160 based platform details IBM System X* 3550 platform with two Intel Xeon processors 5160 3.00GHz, 4Mb L2 cache, 16GB memory (8 x 2GB DDR2-5300F ECC), SUSE* Linux* Enterprise Server 10 (x86_64) 2.6.16.21-0.8-smp, Intel C++ Compiler for Linux* version 10.0. Published in August 2007. Referenced as published at 60.8. For more information see http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070803-01638.html. Intel® Xeon® processor X5470 based platform details Fujitsu Siemens PRIMERGY* RX200 S4 server platform with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (8x2GB DDR2 PC25300F, 2 rank, CAS 5-5-5, with ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 11.0 build 20080730. Referenced as published at 140. For more information see http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080901-05156.html. Intel® Xeon® processor X5570 based platform details ASUS Z8PE-D12X based server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4GB PC3-10600R, CL9-9-9, ECC), SUSE Li Linux E Enterprise t i S Server 10 SP2 x86_64 86 64 K Kernell 2 2.6.16.60-0.34-smp, 6 16 60 0 34 I t l C++ Compiler Intel C il ffor Li Linux32 32 and d Li Linux64 64 version i 11 11.0 0b build ild 20090131. 20090131 R Referenced f d as published bli h d att 241. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090316-06703.html Quad-Core AMD Opteron* processor model 2360 based platform details: Acer Altos G5450 Server platform with two Dual-Core AMD Opteron processors 2360SE 2.50GHz, 2MB L3 cache, memory 32GB (8x4GB, DDR2-667, Reg, 2-rank CL5), SUSE Linux Enterprise Server 10 64-bit, PGI Server Complete Version 7.2 PathScale Compiler Suite 3.2, . Referenced as published at 94.1 For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090309-06627.html Quad-Core AMD Opteron* processor model 2386SE based platform details Supermicro A+ Server 1021M-UR+B* 1021M-UR+B Server platform with two Quad-Core AMD Opteron processors 2386SE 2 2.80GHz, 80GHz 6MB L3 cache cache, 32GB memory (8x4GB (8x4GB, DDR2-800, DDR2-800 Reg, Reg dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 116. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06302.html. Quad-Core AMD Opteron* processor model 2389 based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two Quad-Core AMD Opteron processors 2389 2.90GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 117. For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090216-06529.html

• •

• •

• •

• •

• •

• •

• •





Quad-Core AMD Opteron* processor model 2393 based platform details Dell Inc. PowerEdge SC1435 (AMD Opteron 2393 SE, 3.10 GHz) 148 124 8 2 4 with two Quad-Core AMD Opteron processors 2393SE 3.10GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 124 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07450.html Six-Core Opteron 2434 based platform details

Supermicro A+ Server 1021M-UR+B* Server platform with two six-core AMD Opteron processors 2434 2.6GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Red Hat Enterprise Linux Server release 5.3, PGI Server Complete Version 8.0 x86 Open64 4.2.2 Compiler Suite (from AMD) publishes at 161 http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07357.html



Source: Published/submitted/approved results March 30, 2009.

85

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point Throughput performance on SPECfp*_rate_base2006



Floating Point Throughput Performance on SPECfp*_rate_base2006 benchmark

• • •

Comparison based on published/submitted results to www.spec.org as of June 9, 2009 Intel® Xeon® processor 3.80 with 2M L2 cache based platform details HP* ProLiant* DL380 G4 server platform with two 64-bit Intel Xeon processor 3.80GHz with 2M L2 cache, 800 FSB, 8GB memory, Microsoft Windows Server 2003* Ent. SP1, Intel® C++ Compiler 9.1 for 32-bit apps, Build 20060323Z Package ID: W_CC_P_9.1.020. Referenced as published at 18.8 (SPECfp*_rate_base2006). For more information see http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00025.html. Intel® Xeon® processor 5160 based platform details HP Proliant DL380 G5 server platform* with two Intel Xeon processor 5160, 3.0 GHz, 4MB L2 cache, 1333MHz system bus, 8GB memory (8x1GB FB DDR2-667), SUSE Linux Enterprise Server 10.1 OS*. Intel C++ Compiler for Linux32* and Linux64* version 10.1. Referenced as published at 45.1. Posted at http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02036.html. I t l® Xeon® Intel® X ® processor X5482 b based d platform l tf d details t il Hewlett-Packard ProLiant DL160 G5p server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz CL5 FB-DIMM), 64-Bit SUSE Linux Enterprise Server 10 SP1, Intel C++ Compiler for Linux32 and Linux64 version 10.1 build 20080730. Referenced as published at 86.4. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05587.html. Intel® Xeon® processor X5570 based platform details ASUS Z8PE-D12X based server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 72 GB (18 X 4 GB PC3-10600R, CL=9, SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. References at 197 For more information see: http://www.spec.org/cpu2006/results/res2009q2/cpu2006 http://www spec org/cpu2006/results/res2009q2/cpu2006-20090511-07354 20090511 07354.html html

• •

• •

• •

• •

• •

• •

• •

• •

Quad-Core AMD Opteron* processor model 2360 based platform details: Acer Altos G5450 Server platform with two Dual-Core AMD Opteron processors 2360SE 2.50GHz, 2MB L3 cache, memory 32GB (8x4GB, DDR2-667, Reg, 2-rank CL5), SUSE Linux Enterprise Server 10 64-bit, PGI Server Complete Version 7.2 PathScale Compiler Suite 3.2, . Referenced as published at 85.4. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090309-06628.html Quad-Core Quad Core AMD Opteron* Opteron processor 2386SE based platform details SuperMicro A+ Server 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2386SE 2.80GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server* Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 107. For more information http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06303.html. Quad-Core AMD Opteron* processor model 2389 based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two Quad-Core AMD Opteron processors 2389 2.90GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 108. For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090216-06530.html Quad-Core AMD Opteron* processor model 2393based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two Quad-Core AMD Opteron processors AMD Opteron 2393 SE, 3.1 GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 112 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090406-06932.html Six-Core AMD Opteron* processor 2384 based platform details Supermicro A+ Server 1021M-UR+B* Server platform with two six-core AMD Opteron processors 2434 2.6GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Red Hat Enterprise Linux Server release 5.3, PGI Server Complete Version 8.0 x86 Open64 4.2.2 Compiler Suite (from AMD) publishes at 129 for more information i f ti see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090511-07358.html htt // / 2006/ lt / 2009 2/ 2006 20090511 07358 ht l

Source: Published/submitted/approved results March 30, 2009.

86

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer “SPEED” performance on SPECint*_base2006



Integer SPEED Performance on SPECint*_base2006 benchmark

• • •

Comparison based on published/submitted results to www.spec.org as of March 30, 2009. Intel® Xeon® processor 3.80 with 2M L2 cache based platform details ProLiant* DL380 G4 server platform with two 64-bit Intel Xeon processor 3.80GHz with 2M L2 cache, 800 FSB, 8GB memory, Microsoft Windows Server 2003* Ent. SP1, Intel® C++ Compiler 9.1 for 32-bit apps,, Build 20060323Z Package ID: W_CC_P_9.1.020. Referenced as published at 11.4. For more information see http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00028.html. Intel® Xeon® processor 5160 based platform details HP ProLiant* DL380 G5 platform with two Intel Xeon processors 5160 3.00GHz, 4Mb L2 cache, 8GB memory, SUSE* Linux* Enterprise Server 10 (x86_64) 2.6.16.21-0.8-smp, Intel C++ Compiler for Linux* version 10.0. Published in August 2007. Referenced as published at 19.7. For more information see http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02044.html. Intel® Xeon® processor X5470 based platform details Dell PowerEdge 2950 III server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (4x4GB DDR2 667MHz FB Dimm), SUSE Linux Enterprise Server 10* SP1 x86_64, kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05450.html. Intel® Xeon® processor X5570 based platform details ASUS RS700-E6 server system with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Referenced as published at 32.1 For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090330-06836.html Quad-Core AMD Opteron* processor 2393 based platform details Dell Inc. PowerEdge M805 with two AMD Opteron2393 SE, 3.10 GHz , 6MB L3 cache, 16 GB (4x4 GB, PC2-6400P CL5) memory, SuSE Linux Enterprise Server 10 (x86_64) SP1, Kernel 2.6.16.46-0.12-smp, PGI Server Complete Version 7.2*, PathScale Compiler Suite Version 3.2*. Referenced as published at 19.1For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090428-07285.html

• •

• •

• •

• •

Source: Published/submitted/approved results March 30, 2009.

87

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point “SPEED” performance on SPECfp*_base2006



Floating g Point Speed p Performance on SPECfp* p _base2006 benchmark

• • •

Comparison based on published/submitted results to www.spec.org as of June 9, 2009. Intel® Xeon® processor 3.80 with 2M L2 cache based platform details HP ProLiant DL380 G4* server platform with two 64-bit Intel Xeon processor 3.80GHz with 2M L2 cache, 800 FSB, 8GB memory, Microsoft Windows Server 2003* Ent. SP1, Intel® C++ Compiler 9.1 for 32-bit apps,, Build 20060323Z Package ID: W_CC_P_9.1.020. Referenced as published at 11.7. For more information see http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00026.html. Intel® Xeon® processor 5160 based platform details HP ProLiant DL380 G5* platform with two Intel Xeon processors 5160 3.00GHz, 4Mb L2 cache, 8GB memory, SUSE* Linux* Enterprise Server 10 (x86_64) 2.6.16.46-0.12-smp, IntelC++ Compiler for Linux version 10.1*. Referenced as published at 18.3. For more information see http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02038.html. Intel® Xeon® processor X5482 based platform details IBM System x 3450* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory, 64-Bit SUSE Linux Enterprise Server 10*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006 20081215 06283.html. http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081215-06283.html. Intel® Xeon® processor X5570 based platform details Hitachi BladeSymphony BS2000 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR31333MHz), RHEL 5.3 x86_64 Kernel 2.6.18-128.el5, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Referenced as published at 39.3. For more information see http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090317-06805.html Quad-Core AMD Opteron* processor model 2389 based platform details HP ProLiant DL385 G5p* platform with two AMD Opteron 2389, 2.9 GHz 6MB L3 cache, 32 GB (8x4 GB, PC2-6400P CL5) memory, Red Hat Enterprise Linux Server release 5.3,Kernel 2.6.18-128.el5 , PGI Server Complete Version 8.0, PathScale Compiler Suite Version 3.2*. Referenced as published at 22.8. For more information see: http://www.spec.org/cpu2006/results/res2009q2/cpu2006-20090427-07257.html

• •

• •

• •

• •

.

Source: Published/submitted/approved results March 30, 2009.

88

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java performance on SPECjbb*2005



Java Performance on SPECjbb*2005 benchmark

• •

IIntel® t l® X Xeon® ® processor 3 3.80 80 with ith 2M L2 cache h based b d platform l tf details d t il Intel® Server Platform with two 64-bit Intel Xeon processor 3.80GHz with 2M L2 cache, 800 FSB, 8x1GB DDR2-400 memory, Microsoft Windows Server* 2003 Ent. SP1, BEA* JRockit* build P27.4.0-windows-x86_64 run with 2 JVM instances. Performance measured at 50,970 BOPS. Intel internal measurement. Intel® Xeon® processor 5160 based platform details Fujitsu Siemens* PRIMERGY* BX620 S3 platform with two Intel Xeon processors 5160 3.00GHz, 4Mb L2 cache, 8x1GB DDR2-RAM PC2-5300F (CAS 5-5-5) memory, Microsoft Windows Server 2003 Enterprise x64 Edition* + SP1 (64-bit), BEA JRockit 5.0 P27.2.0 . 1 JVM instance. Referenced as published at 138,388 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2007q1/jbb2005-20070213-00254.html. Intel® Xeon® processor X5470 based platform details Fujitsu Siemens PRIMERGY RX200 S4 server platform* with two Intel Xeon processors 5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory, Microsoft Windows Server 2008 Enterprise x64 Edition*, Oracle JRockit 6 P28.0.0 (build P28.0.0-8-109238-1.6.0_05-20090130-1408-windows-x86_64) 4 JVM instances. Referenced as published at 368,034 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090220-00583.html. Intel® Xeon® processor X5460 based platform details Sun Fire X4150 server platform* with two Intel Xeon processors 5460 3.16 GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory, Solaris 10 8/07 OS*, Java HotSpot(TM) 32-Bit Server VM on Solaris, version 1.6.0_05-p 4 JVM instances. Referenced as published at 303,297 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2008q1/jbb200520080208-00452.html . Intel® Xeon® processor X5470 based platform details IBM System x3650 server platform* with two Intel Xeon processors 5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory, Microsoft Windows Server 2003 Enterprise x64 Edition*, IBM J9 2.4 JRE 1.6.0 (build pwa6460sr3-20080924_03) 4 JVM instances. Referenced as published at 344,436 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081031-00555.html Intel® Xeon® processor X5570 based platform details

• •

• •

• •

• •

• •

Dell Inc. PowerEdge T610 with two Intel Xeon X5570, 2.93 GHz),8 cores total, HT on, Memory (MB) 49152, Memory Details 12 x 4 GB DDR3-1333 DR RDIMM, Primary cache 32 KB I + 32 KB D on chip per core,Secondary cache 256 KB I+D on chip per core,Other cache 8 MB I+D L3 on chip per chip, Oracle JRockit(R) 6 P28.0.0 (build P28.0.0-29-114096-1.6.0_1120090427-1759-windows-x86 64 compiled mode). 20090427-1759-windows-x86_64, mode) For more information see: http://www.spec.org/osg/jbb2005/results/res2009q2/jbb2005-20090519-00727.html http://www spec org/osg/jbb2005/results/res2009q2/jbb2005-20090519-00727 html

• •

Intel® Xeon® processor X5570 based platform details Sun Fire X4270 Server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Microsoft Windows Server 2008 Enterprise x64 Edition, Java HotSpot(TM) 32-Bit Server VM on Solaris, version 1.6.0_14. 2 JVM instances. Result measured at 556,822 BOPS - March 30, 2009. Intel® Xeon® processor X5570 based platform details IBM Bladecenter HS22 Server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Microsoft Windows Server 2008 Enterprise x64 Edition, IBM J9* 2.4 JRE 1.6.0 (build pwa6460sr5-20090323_04(SR5)) run with 4 JVM instances. Result submitted to www.spec.org for review at 604417 BOPS as of March 30,, 2009.

• •

• •

Quad-Core AMD Opteron* processor 2393SE based platform details SuperMicro A+ Server 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2393SE 3.1 GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Microsoft Windows Server 2008 Enterprise with SP1 (64-bit), IBM J9 JVM build 2.4, J2RE 1.6.0 IBM J9 2.4. 2 JVM instances. Referenced as published at 408202 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q2/jbb2005-20090406-00704.html



Six-Core AMD Opteron* processor 2434 based platform details



SuperMicro A+ Server 1021M-UR+B Server platform* with two six-core AMD Opteron processors 2434 2.6 GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Microsoft Windows Server 2008 Enterprise with SP1 (64-bit), IBM J9 JVM build 2.4, J2RE 1.6.0 IBM J9 2.4. 2 JVM instances. Referenced as published at 462189 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q2/jbb2005-20090505-00717.html

Source: Published/submitted/approved results June 2, 2009.

89

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Database performance on TPC*-E



Database performance on TPC-E* benchmark

• • •

Comparison C i b based d on published/submitted bli h d/ b i d results l at www.tpc.org as off March M h 30, 30 2009 2009. Intel® Xeon® processor 5160 based platform details IBM Bladecenter HS21 XM Server platform* with two Intel Xeon processor 5160 3.0GHz (2 processors / 4 cores / 4 threads), 4MB L2 cache, 32GB memory, Microsoft SQL Server* 2005 x64 Enterprise Edition SP2*, Microsoft Windows Server* 2003 Enterprise x64 Ent. SP2*. Referenced as published at 169.59 tpsE and $1,897.66/tpsE; availability date Aug 10, 2007. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=2. Intel® Xeon® processor X5460 based platform details Fujitsu Siemens PRIMERGY TX300 S4 server platform* Fujitsu-Siemens platform with Intel Xeon processor X5460 3 3.16GHz 16GHz (2 processors / 8 cores / 8 threads) threads), 2x6MB L2 cache, cache 1333 MHz system bus, 64GB memory, Microsoft SQL Server 200 x64 Enterprise Edition*, Microsoft Windows Server 2008 Enterprise x64*. Referenced as published at 317.45 tpsE and $523.49/tpsE; availability date August 30, 2008. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=12. Intel® Xeon® processor X5570 based platform details Fujitsu-Siemens PRIMERGY RX300 S5 server platform* with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066), Microsoft SQL Server 2008 x64 Enterprise Edition, Microsoft Windows Server 2008 Enterprise x64. Referenced as published at 800tpsE and $343.91/tpsE. Availability date April 1, 2009. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=25

• •

• •

Source: Published/submitted/approved results June 2, 2009.

90

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Database performance on TPC*-C



Database performance on TPC-C* benchmark

• • •

Comparison b C based d on published/submitted bl h d/ b d results l at www.tpc.org as off March h 30, 30 2009. 2009 Intel® Xeon® processor 3.60 with 2M L2 cache based platform details HP ProLiant ML370-G4* server platform with Two 64-bit Intel Xeon processors 3.60GHz (2 processors / 2 cores / 4 threads) with 2MB L2 cache 800 MHz system bus, 16 GB DDR2, Microsoft Windows* Server 2003 Enterprise Edition, Microsoft SQL* Server 2000 Enterprise Edition SP3. Results at http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=105021101 Referenced as published: 74,298 tpmC; $2.40/tpmC; Availability Date Feb 14, 2005. Intel® Xeon® processor 5160 based platform details HP ProLiant ML370 G5 G5* SAS 3 3.0 0 GHz/4MB platform with Intel Xeon processor 5160 3.0GHz 3 0GHz (2 processors / 4 cores / 4 threads) threads), 4MB L2 cache cache, 8GB memory, memory Microsoft SQL Server 2005 x64 Enterprise Edition SP1*, Microsoft Windows Server 2003 Enterprise x64 Ent. R1*. Referenced as published at 169,360 tpmC and $2.93/tpmC; availability date Nov 22, 2006. For more information see http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=106052202. Intel® Xeon® processor X5460 based platform details HP ProLiant ML370 G5 platform with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Oracle Database 10g Standard Edition One with Oracle Enterprise Linux OS*. Referenced as published at 273,666 tpmC and $1.38/tpmC; availability date November 12, 2007. For more information see http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=107111201 Intel® Xeon® processor X5460 based platform details HP ProLiant ML370 G5 platform with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Microsoft SQL Server 2005 x64 Enterprise Edition SP2, Microsoft Windows Server 2003 Enterprise x64 Ent. R2. Referenced as published at 275,149 tpmC and $1.44/tpmC; availability date January 7, 2008. For more information see http://tpc.org/tpcc/results/tpcc_result_detail.asp?id=108010701. Intel® Xeon® processor X5570 based platform details HP ProLiant DL370 G6* platform with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 144 GB memory (18x8 GB DDR3). DDR3) Oracle Database 11g Standard Edition One * with Oracle Enterprise Linux OS OS*. Referenced as published at 631 631,766 766 tpmC and $1 $1.08/tpmC; 08/tpmC; availability date March 30, 2009. For more information see http://www.tpc.org/tpcc/results/tpcc_result_detail.asp?id=109033001

• •

• •

• •

• •

Source: Published/submitted/approved results June 2, 2009.

91

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Web Server performance on SPECWeb*2005



Web server performance using SPECweb2005

• • •

Comparison C i b based d on published/submitted bli h d/ b i d results l to www.spec.org as off March M h 30 30, 2009 2009. Intel® Xeon® processor 3.80 with 2M L2 cache based platform details Fujitsu Siemens PRIMERGY* TX300 S2 server platform with two Intel Xeon processors 3.80GHz, 2MB L2 cache, 16GB SDRAM, RedHat Enterprise* Linux* 4 2.6.922 ELsmp x86_64, Rock Web Server* v1.3 x86_64. Referenced as published at 4,555. For more information see http://www.spec.org/web2005/results/res2006q1/web2005-20060130-00021.html. Intel® Xeon® processor 5160 based platform details HP ProLiant ProLiant* ML360 G5 server platform with two Intel Xeon processors 5160 3.00GHz, 3 00GHz 4MB L2 cache cache, 32GB memory (8x4GB 667MHz ECC CL5 DDR2 FB-DIMM), FB DIMM) RedHat Enterprise Linux 5 2.6.18-8.e15, Rock Web Server v1.4.1 (x86_64). Referenced as published at 15,193. For more information see http://www.spec.org/web2005/results/res2007q2/web2005-20070604-00068.html. Intel® Xeon® processor X5460 based platform details HP Proliant DL380 G5 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 32GB memory (8x4G 667MHz ECC DDR2 FB-DIMM), RedHat Enterprise Linux 5 (2.6.18-53.el5), Rock Web Server v1.4.6 x86_64. Referenced as published at 29591. For more information see http://www.spec.org/web2005/results/res2008q1/web2005-20080225-00104.html. Intel® Xeon® processor X5570 based platform details HP ProLiant DL380 G6 platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 144 GB memory (18x8 GB DDR3), Red Hat Enterprise Linux 5.2 .Rock Web Server v1.4.7 (x86_64). Result published at 75023. For more information see http://www.spec.org/osg/web2005/results/res2009q2/web2005-20090423-00135.html

• •

• •

• •

• •

• •

Quad-Core AMD Opteron* processor model 2360 based platform details: Sun Fire X4240 Server platform with two Dual Dual-Core Core AMD Opteron processors 2360SE 2 2.50GHz, 50GHz 2MB L3 cache cache, memory 32GB (16x2GB (16x2GB, DDR2 DDR2-667), 667) RedHat* Enterprise Linux .5.2 (2.6.18-91.el5), Rock Web Server v1.4.7 x86_64. Referenced as published at 32288. For more information see http://www.spec.org/osg/web2005/results/res2008q4/web2005-20081022-00117.html Quad-Core AMD Opteron* processor model 2384 based platform details HP Proliant DL165 G5 Server platform with two Dual-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, memory 32GB (8x4GB, PC2-6400P CL5), RedHat* Enterprise Linux .5.2 (2.6.18-92.el5), Rock Web Server v1.4.7 x86_64. Referenced as published at 39530. For more information see http://www.spec.org/osg/web2005/results/res2009q1/web2005-20090122-00125.html.

Source: Published/submitted/approved results June 2, 2009.

92

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Enterprise Resource Planning performance on SAP-SD* 2-Tier



Enterprise Resource Planning Performance on SAP-SD* 2-Tier benchmark

• • •

All results l published bli h d at http://www.sap.com/solutions/benchmark/sd2tier.epx h // / l i /b h k/ d2 i as off March M h 30, 30 2009 2009. Intel® Xeon® processor 3.60 GHz based platform details HP* ProLiant* ML370 G4 Model HPS, 2-way SMP, 64-bit Intel Xeon processor 3.60GHz with 2MB L2 cache 800 MHz system bus, 16 GB main Microsoft Windows* Server 2003 Enterprise Edition (64-bit), Microsoft SQL* Server 2000 (32-bit), SAP SD* standard R/3 Enterprise 4.70 application benchmark. Referenced as published at 597 SD users. Certification number 2005005. Intel® Xeon® processor 5160 based platform details IBM System* System x3650 server platform with two Intel Xeon processors 5160 5160, 3.00GHz 3 00GHz with 4M L2 cache cache, 1333MHz system bus bus, 24GB FBDIMM memory, memory Microsoft Windows Server 2003 Enterprise Edition, IBM DB2* UDB 8.2.2 database, SAP ECC* Release 5.0. Referenced as published at 1,283 SD users. Certification number 2006043. Intel® Xeon® processor X5470 based platform details HP ProLiant BL460C server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005, SAP ECC Release 6.0 . Referenced as published at 2,518 SD users. Certification number 2008048. Intel® Xeon® processor X5570 based platform details Sun Fire X4270 Server with two Intel Xeon processors X5570, 2.93GHz 8MB L3 cache, 6.4QPI, 48GB memory, Solaris 10, Oracle 10g, SAP ECC Release 6.0 Enhancement Pack 4 (Unicode) . Referenced as published at 3,700 SD users. Certification number 2009005. Intel® Xeon® processor X5570 based platform details HP Proliant DL380 G6p Server platform with two Intel Xeon processors X5570, 2.93GHz 8MB L3 cache, 6.4QPI, 48GB memory, Microsoft Windows Server 2003 Enterprise Edition, DB2 9.5, SAP ECC Release 6.0 (2005). Referenced as published at 5,100 SD users. Certification number 2008079. Intel® Xeon® processor X5570 based platform details IBM S System t x3650 3650 M2 S Server with ith ttwo IIntel t lX Xeon processors X5570 X5570, 2.93GHz 2 93GH 8MB L3 cache, h 6.4QPI, 6 4QPI 48GB memory, Mi Microsoft ft Windows Wi d S Server 2003 E Enterprise t i Edition, SQL Server 2005, SAP ECC Release 6.0 (2005). Referenced as published at 4,995 SD users. Certification number 2008071. Quad-Core AMD Opteron* processor model 2384 based platform details HP Proliant DL385 G5p Server platform* with two Quad-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005 database, SAP ECC Release 6.0. Referenced as published at 2,752 SD Users. Certification number 2008065. Six-Core AMD Opteron* processor model 2434 based platform details HP Proliant DL385 G5p Server platform platform* with two Six-Core AMD Opteron processors 2434 2434, 2 2.60GHz, 60GHz 6MB L3 cache cache, 32GB memory memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005 database, SAP ECC Release 6.0 Enhancement Pack 4 (Unicode) , Referenced as published at 2355 SD Users. Certification number 2009020

• •

• • • • • • • • • • • •

Source: Published/submitted/approved results June 2, 2009.

93

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java Application and Application Server performance



SPECjAppServer2004

• • •

Comparison based on published/submitted results to www.spec.org as of June 2, 2009. Baseline Intel® Xeon® processor X5460 based platform details: HP Proliant BL460c G1 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 16GB memory (8x2G 667MHz ECC DDR2 FB-DIMM), Oracle Application Server 10G Release 10.1.3.3 - Java Edition, BEA JRockit(R) 6.0 JDK (R27.3.0-106) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.6. Referenced as published at 2056. For more information see http://www.spec.org/osg/jAppServer2004/results/res2007q4/jAppServer2004-2007102300088.html . Intel® Xeon® processor X5570 based platform details: Dell PowerEdge R610 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (12x2 GB DDR3), Oracle WebLogic Server Standard Edition Release 10.3, Oracle JRockit(R) 6.0 JDK (R27.6.0-50) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.7, Result submitted to www.spec.org for review at 3975 as of March 30, 2009.

• •



SPECjvm2008:

• • •

Comparison based on published/submitted /measured results as of June 2, 2009. Intel® Xeon® processor X5470 based platform details: Supermicro X7DBE +* server platform with two Intel® Xeon® processors X5470 3.333GHz, 12MB L2 Cache, 1333MHz FSB, 16GB memory (8x2GB 667MHz DDR2 FB-DIMM), Red Hat Enterprise Linux 5.3 (kernel: 2.6.18-128.el5), JDK 1.6.0_06-p-b04. Intel internal measurement. Measured at 179.53 SPECjvm2008 base ops/m as of March 30, 2009 Intel® Xeon® processor X5570 based platform details: Sun Blade X6270* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR3-1333MHz), OpenSolaris* 2008 11 ZFS, 2008.11, ZFS JDK 1.6.0_14. 1 6 0 14 Result submitted to www.spec.org www spec org for review at 317 317.13 13 SPECjvm2008 base ops/m as of April 14 14, 2009 2009.

• •

Source: Published/submitted/approved results June 2, 2009.

94

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficient performance on SPECpower*_ssj2008

Perf/Watt Perf /Watt



Energy efficient performance using SPECpower*_ssj2008



Intel® Xeon® processor 3.60 with 1M L2 cache based platform details



Intel® Server Board SE7520AF2 (3.6 GHz/1M L2 Intel® Xeon® processor) server platform with two Intel Xeon processors 3.6 GHz, 1MB L2 cache, 4 GB memory (4x1), BEA JRockit* 1.6.0_02 (build P27.4.0-10-90053-1.6.0_02-20071009-1827-windows-x86_64, compiled mode). Published at 87.4 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2007q4/power_ssj2008-20071129-00015.html.

• •

Intel® Xeon® processor 5160 based platform details Supermicro 6025B-TR+ (Intel Xeon processor 5160) server platform* with two Intel Xeon processors 5160 3.0 GHz, 4MB L2 cache, 8 GB memory (4x2), BEA JRockit* 1.6.0_02 (build P27.4.0-10-90053-1.6.0_02-20071009-1827-windows-x86_64, compiled mode). Published at 338 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2007q4/power_ssj2008-20071129-00016.html.

• •

processor L5430 based platform p details Intel® Xeon® p Fujitsu PRIMERGY TX300 S4 server platform* with two Intel Xeon processors L5430 2.33GHz, 12MB L2 cache, 1333MHz FSB, 8GB memory, Oracle JRockit(R) 6 P28.0.0 (build P28.0.0-8-109238-1.6.0_05-20090130-1408-windows-x86_64). Published at 917 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2009q1/power_ssj2008-20090220-00124.html

• •

Intel® Xeon® processor L5430 based platform details Powerleader PR2510D2 server platform* with two Intel Xeon processors L5430 2.33GHz, 12MB L2 cache, 1333MHz FSB, 8GB memory, Oracle JRockit* (build P27 5 0 5 P27.5.0-5_o_CR371811_CR374296-100684-1.6.0_03-20080702-1651-windows-x86_64, CR371811 CR374296 100684 1 6 0 03 20080702 1651 i d 86 64 compiled il d mode). d ) P Published bli h d att 1135 ssj_ops/watt. j / tt For F more information i f ti see: http://www.spec.org/power_ssj2008/results/res2008q4/power_ssj2008-20081007-00086.html.

• •

Intel® Xeon® processor X5570 based platform details IBM System x3650 M2* server platform with two Intel Xeon processor X5570, 2.93GHz, 8 GB (4 x 2) memory, Microsoft Windows Server 2008 Enterprise* OS. IBM J9 Java™6 Runtime Environment JVM. Result submitted to www.spec.org at 1977 ssj_ops/watt for more information see: http://www.spec.org/power ssj2008/results/res2009q2/power ssj2008-20090519-00165.html http://www.spec.org/power_ssj2008/results/res2009q2/power_ssj2008 20090519 00165.html

• •

Quad-Core AMD Opteron* processor model 2377EE based platform based details ZT Systems, 1224Ra Datacenter Server platform with two AMD Opteron 2377 EE, Quad-Core, 2.3GHz, 6MB L3 Cache, 4X4GB, Windows Server 2008 ,IBM J9 VM (build 2.4, J2RE 1.6.0 IBM J9 2.4 amd64-64 jvmwa6460-20081105_25433 (JIT enabled, AOT enabled)) result published at 1379 for more information see: http://www.spec.org/power_ssj2008/results/res2009q2/power_ssj2008-20090504-00149.html

Source: Published/submitted/approved results June 2, 2009.

95

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficient performance on SPECint*_rate and SPECfp*_rate

Perf/Watt Perf /Watt



Energy efficient performance using SPECint*_rate_base2006



Comparison based on published results by Principled Technologies at http://www.principledtechnologies.com/Clients/Reports/Intel/X5570.htm as of March 30, 2009. Pl tf Platform power measured d during d i steady t d state t t off performance f run to t calculate l l t Perf/Watt. P f/W tt Intel® Xeon® processor 3.60 with 2M L2 cache based platform details Supermicro X6DHT-G server platform* with two 64-bit Intel Xeon processor 3.60GHz with 2M L2 cache, 800 FSB, 16GB (8x2GB DDR-333MHz) memory, SLES* 10 SP2 OS, (2.6.16.60-0.21-smp Kernel*). Spec binaries built with Intel® C/C++ Compiler 9.1.053 for Extended Memory 64 Technology Edition. Intel® Xeon® processor E5450 based platform details Supermicro X7DBE+ server platform* with two Intel Xeon processors E5450 3.0GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (8x2GB DDR2 FBDimm 667), SLES 10 SP2 OS, OS (2.6.16.60 (2 6 16 60-0 0.21 21-smp smp Kernel) Kernel). Spec binaries built with Intel C/C++ Compiler 11 11.0.074 0 074 for Extended Memory 64 Technology Edition. Edition Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4GB DDR3 1333), SLES 10 SP2 OS, (2.6.16.60-0.21-smp Kernel). Spec binaries built with Intel C/C++ Compiler 11.0.080 for Extended Memory 64 Technology Edition. Turbo Mode disabled. Quad-Core AMD Opteron* processor 2376HE based platform based details Supermicro H8DME-2 Server platform* with two Quad-Core AMD Opteron processors 2376HE 2.70GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800), SLES 10 SP2 OS, (2.6.16.60-0.21-smp Kernel). Spec binaries built with PGI Server* Complete Version 8.0-3 PathScale Compiler Suite*, Release 3.2.

• • • • • • • •



Energy efficient performance using SPECfp*_rate_base2006



Comparison based on published results by Principled Technologies at http://www.principledtechnologies.com/Clients/Reports/Intel/X5570.htm as of March 30, 2009. Platform power measured during steady state of performance run to calculate Perf/Watt. Intel® Xeon® processor 3.60 with 2M L2 cache based platform details Supermicro X6DHT X6DHT-G G server platform platform* with two 64-bit 64 bit Intel Xeon processor 3.60GHz with 2M L2 cache, 800 FSB, 16GB (8x2GB DDR-333MHz) DDR 333MHz) memory, SLES SLES* 10 SP2 OS, (2.6.16.60-0.21-smp Kernel*). Spec binaries built with Intel® C/C++ Compiler 9.1.053 for Extended Memory 64 Technology Edition. Intel® Xeon® processor E5472 based platform details Supermicro X7DWN+ server platform* with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB DDR2 FBDimm 667), SLES 10 SP2 OS, (2.6.16.60-0.21-smp Kernel). Spec binaries built with Intel C/C++ Compiler 11.0.074 for Extended Memory 64 Technology Edition. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4GB DDR3 1333), SLES 10 SP2 OS, (2.6.16.60-0.21-smp Kernel). Spec binaries built with Intel C/C++ Compiler 11.0.080 for Extended Memory 64 Technology Edition. Turbo Mode disabled. Quad-Core AMD Opteron* processor 2376HE platform based details Supermicro H8DME-2 Server platform* with two Quad-Core AMD Opteron processors 2376HE 2.70GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800), SLES 10 SP2 OS, (2.6.16.60-0.21-smp Kernel). Spec binaries built with PGI Server* Complete Version 8.0-3 PathScale Compiler* Suite, Release 3.2.

• • • • • • • •

Source: Published/submitted/approved results March 30, 2009.

96

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Virtualization performance on VMware ESX* using VMmark* benchmark •

Virtualization Performance on VMware* ESX using VMmark* benchmark

• • •

Comparison C i b based d on published/submitted/approved bli h d/ b itt d/ d results lt as off A Aprilil 27 27, 2009 2009. R Results lt are posted t d att http://www.vmware.com/products/vmmark/results.html htt // / d t / k/ lt ht l Intel® Xeon® processor 5160 based platform details Dell PowerEdge* 2950 server platform with two Intel Xeon processors 5160 3.00Ghz, 4MB L2 cache, 1333MHz FSB, 32GB memory (8x4GB 667MHz ECC CL5 DDR2 FB-DIMM), VMware ESX Server* V3.0.1. Published at 3.89@3 tiles. For more information see www.vmware.com/files/pdf/vmmark_dell3.PDF Intel® Xeon® processor X5470 based platform details HP Proliant* ML370 G5 server platform with two Intel Xeon processors X5470 3.33GHz, 2x6MB L2 cache, 1333MHz FSB, 48GB memory, VMware ESX V3.5. Update 3 Published at 9.15@7 tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-10-09-ML370.pdf I t l® Xeon® Intel® X ® processor X5570 b based d platform l tf d details: t il Inspur NF5280 Server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 23.45@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-Inspur-2009-03-30-NF5280.pdf Intel® Xeon® processor X5570 based platform details IBM* System x* 3650 M2 Server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 23.89@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-IBM-2009-04-27-3650M2.pdf Intel® Xeon® processor X5570 based platform details: Dell PowerEdge* R710 Server platform with two Intel Xeon processors X5570 2 2.93GHz, 93GHz 8MB L3 cache cache, 6 6.4GT/s 4GT/s QPI QPI, 96 GB memory (12x8 GB DDR3-1066MHz), DDR3 1066MHz) VMware ESX v4.0. VMmark V1.1. Result measured at 24.0@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-Dell-2009-04-21-R710.pdf Intel® Xeon® processor X5570 based platform details: Cisco B-200 M1 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 24.14@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-Cisco-2009-04-21-B200M1.pdf HP ProLiant BL490 G6 VMware ESX v4.0 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX v4.0. VMmark V1.1. Result measured at 24.24@17 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-HP-2009-05-19-BL490G6.pdf AMD Opteron* Opteron processor model 2360 based platform details Sun Fire X4240 platform with two AMD Opteron 2360SE, 2MB L3 Cache, 32GB memory. VMware ESX V3.5. Update 1 Published at 8.07@6 tiles. For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-Sun-2008-10-23-X4240.pdf . AMD Opteron* processor model 2384 based platform details HP ProLiant* DL385 G5p platform with two AMD Opteron 2384, 6MB L3 Cache, 32GB memory. VMware ESX V3.5. Update 3 Published at 11.28@8 tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-12-30-DL385.pdf. Six-core AMD Opteron* processor model 2434 based platform details HP ProLiant ProLiant* DL385 G5p platform with two AMD Opteron 2434, 2434 6MB L3 Cache, Cache 64 GB memory. memory VMware ESX V4 V4.0 0 published at 15 15.54@11 54@11 tiles. tiles For more information see http://www.vmware.com/files/pdf/vmmark/VMmark-HP-2009-06-02-dl385g6.pdf

• • • • • • • • • • • • • • • • •

Source: Published/submitted/approved results June 9,2009

97

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Legacy Database and Mail Server Consolidation under Virtualization with Microsoft Hyper-v •

Legacy Database Server Consolidation Using Microsoft Hyper-V*



Comparison based on published results by Principled Technologies at http://www.principledtechnologies.com/Clients/Reports/Intel/X5570.htm as of March 30, 2009. Microsoft SQLServer2005 was used as the database server. Note that different platform sustains a different average score but they are all above the score of the baseline platform. Intel® Xeon® processor 3.60 with 2M L2 cache based platform details Supermicro X6DHT-G server platform with two 64-bit Intel Xeon Processor 3.60Ghz with 2M L2 Cache, 800 FSB, 4GB (2x2GB DDR-333MHz) memory, Microsoft Windows Server 2003 R2 Enterprise Edition. Intel® Xeon® processor E5450 based platform details Supermicro X7DBE+ server platform with two Intel Xeon processors E5450 3.0GHz, 12MB L2 cache, 1333MHz FSB, 64GB memory (16x4GB DDR2 FBDimm 667), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN + server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 72 GB memory (18x4GB DDR3 800), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V. AMD Opteron* processor model 2384 based platform details Supermicro H8DME-2 H8DME 2 Server platform with two Quad Quad-Core Core AMD Opteron processors 2384 2 2.70GHz, 70GHz 6MB L3 cache cache, 64GB memory (16x4GB DDR2 667) 667), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V.

• • • • • • • •



Legacy Mail Server Consolidation Using Microsoft Hyper-V*



Comparison based on published results by Principled Technologies at http://www.principledtechnologies.com/Clients/Reports/Intel/X5570.htm as of March 30, 2009. Microsoft Exchange Server was used as the Mail server. Note that different platform has a different CPU utilization level. Intel® Xeon® processor 3.60 3 60 with 2M L2 cache based platform details Supermicro X6DHT-G server platform with two 64-bit Intel Xeon Processor 3.60Ghz with 2M L2 Cache, 800 FSB, 4GB (2x2GB DDR-333MHz) memory, Microsoft Windows Server 2003 R2 Enterprise Edition. Intel® Xeon® processor E5450 based platform details Supermicro X7DBE+ server platform with two Intel Xeon processors E5450 3.0GHz, 12MB L2 cache, 1333MHz FSB, 64GB memory (16x4GB DDR2 FBDimm 667), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN + server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 72 GB memory (18x4GB DDR3 800), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V. AMD Opteron* processor model 2384 platform based details Supermicro H8DME-2 Server platform with two Quad-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, 64GB memory (16x4GB DDR2 667), Windows Server 2008, Enterprise x64 Edition with Microsoft Hyper-V.

• • • • • • • •

Source: Published/submitted/approved results June 2, 2009.

98

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Open MP performance on SPECOMP*2001 – Medium and Large Workloads • •

OpenMP* and shared-memory systems performance using SPEC OMP2001* Comparison based on published/submitted results to www.spec.org as of March 30, 2009.

SPEComp*Mbase2001: SPECompM results on Intel® Xeon® processor 5500 series-based server were obtained using 16 OpenMP threads over two sockets, on Intel® Xeon® processor 5400 series and AMD Opteron 2300 series-based server 8 OpenMP threads over two sockets were used. Xeon 5100 series is based on 4 OpenMP threads. • •

• •

• •

• •

Intel® Xeon® processor 5160 based platform details Intel® Server Board S5000XAL platform with two Intel Xeon processors 5160 3.0GHz, 4MB L2 cache, 1333MHz FSB, 8GB memory (8x1GB 667MHz DDR2 ) Red Hat Enterprise p Linux* AS release 4 ((Nahant Update p 4)) Binaries built with Intel® Compiler p 9.1. Referenced as p published at 10689. FB-DIMM), (SPECompMbase2001). For more information see http://www.spec.org/omp/results/res2006q4/omp2001-20061120-00230.html. Intel® Xeon® processor E5472 based platform details Supermicro X7DB8+ server platform* with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory (8x4GB 800MHz DDR2 FB-DIMM), SUSE LINUX 10.1* (X86-64) (Linux 2.6.16.13-4-smp). Binaries built with Intel Compiler 10.1. Referenced as published at 17187. (SPECompMbase2001). For more information see http://www.spec.org/omp/results/res2007q4/omp2001-20071107-00274.html. Intel® Xeon® processor X5570 based platform details S Sun Blade Bl d X6275 server platform* l tf * with ith ttwo IIntel t lX Xeon processors X5570 2.93GHz, 2 93GH 8MB L3 cache, h 6.4GT/s 6 4GT/ QPI, QPI 24 GB memory (6 (6x4 4 GB DDR3 DDR31333MHz), OpenMP and Automatic parallel Operating System: OpenSolaris 2008.11 Compiler: Sun Studio 12 Update 1 (backend build 20090309) Result published at 43777 (SPECompMbase2001) for more information see: http://www.spec.org/omp/results/res2009q2/omp2001-20090413-00350.html Quad-Core AMD Opteron processor 2384 based platform* details Supermicro H8DMU Server platform* with two Quad-Core AMD Opteron processors 2386SE* 2.80GHz, 6MB L3 cache, 16GB memory (8x2GB, PC2-6400, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit, Binaries built with PathScale Compiler Suite*, Release 3.1. Referenced as published at 22678 (SPECompMbase2001). For more information http://www.spec.org/omp/results/res2008q4/omp2001-20081021-00320.html.



SPEComp*Lbase2001: SPECompL results on Intel® Xeon® processor 5500 series-based server were obtained using 16 OpenMP threads over two sockets, on Intel® Xeon® processor 5400 series-based server 8 OpenMP threads over two sockets were used.

• •

Baseline Intel® Xeon® processor X5482 based platform details: Supermicro p X7DB8+ server p platform* with two Intel Xeon p processors E5472 3.0GHz,, 12MB L2 cache,, 1600MHz FSB,, 32GB memoryy (8x4GB ( 800MHz DDR2 FB-DIMM), SUSE LINUX 10 SP2. Binaries built with Intel Compiler 11.0. Result submitted to www.spec.org for review at 81332 (SPECompLbase2001) as of March 30, 2009. Intel® Xeon® processor X5570 based platform details: Sun Blade X6275 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR31333MHz), OpenMP and Automatic parallel Operating System: OpenSolaris 2008.11 Compiler: Sun Studio 12 Update 1 (backend build 20090309) Result published at 237781 (SPECompLbase2001) for more information see: http://www spec org/omp/results/res2009q2/omp2001-20090413-00349 http://www.spec.org/omp/results/res2009q2/omp2001 20090413 00349.html html

• •

Source: Published/submitted/approved results June 2, 2009.

99

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms HPC -ANSYS* FLUENT* and LS-DYNA*



Computational Fluid Dynamics analysis using ANSYS* FLUENT (Single Node)



Comparison based on published/submitted results to www.fluent.com/software/fluent/fl6bench/fl6bench_6.4.x/index.htm www.fluent.com/software/fluent/fl6bench/fl6bench 6.4.x/index.htm as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.13)*. Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchmarks. Intel® Xeon® processor X5570 based platform details SGI Altix ICE 8200EX* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.9) Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchmarks. Quad-Core AMD Opteron* processor model 2384 platform based details Server platform with two AMD Opteron 2384 processor 2.7GHz, 6MB L3 cache, Linux OS. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.7) Six individual benchmarks are shown as a measure of single node performance. "Overall" performance is the geometric mean of the six individual benchmarks.

• •

• •

• •



Crash Simulation analysis using LS-DYNA* (Single Node)



Comparison based on published/submitted results to http://www.topcrunch.org/ as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8 X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux* 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car.

• •

• •

Source: Published/submitted/approved results March 30, 2009.

100

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms HPC -Abaqus*, ANSYS*Mechanical* and Star-CD*



Unified Finite Element Analysis using Abaqus*

• • •

Comparison based on published/submitted results to www.simulia.com/support/v68/v68_performance.php as of March 30, 2009. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Abaqus/Standard* and Abaqus/Explicit v6.8*-EF1. Benchmark for Abaqas/Standard consists of a suite of 10 workloads and Abaqas/Explicit consists of a suite of 6 workloads. Geo mean of each these workload groups used for comparison. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit 64 bit RedHat Enterprise Linux 5 5.3. 3 Performance measured using Abaqus/Standard and Abaqus/Explicit v6.8 v6 8-EF1 EF1. Benchmark for Abaqas/Standard consists of a suite of 10 workloads and Abaqas/Explicit consists of a suite of 6 workloads. Geo mean of each these workload groups used for comparison.

• •



Multiphysics Finite Element Analysis using ANSYS* Mechanical*

• • •

Comparison based on published/submitted results to www.ansys.com/services/hardware-support-db.htm as of March 30, 2009. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+ X7DB8+* server platform with two Intel® Xeon® processors X5482 3 3.20GHz, 20GHz 12MB L2 cache cache, 1600MHz FSB FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.1*. Performance measured using ANSYS* Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared* consists of a suite of 8 workloads and Ansys-Distributed* consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using ANSYS* Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared consists of a suite of 8 workloads and Ansys-Distributed consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison.

• •



Computational Fluid Dynamics analysis using Star-CD*



Comparison based on published/submitted results to http://www.cd-adapco.com/products/STAR-CD/performance/406/index.html as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM) 64 DIMM), 64-bit bit RedHat R dH t Enterprise E t i Li Linux 5 5.3*. 3* Performance P f measured d using i STAR-CD STAR CD v4.06. 4 06 Same S configuration fi ti used d ffor allll b both th b benchmark h k results lt - A-Class A Cl and d C-Class. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI* 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using STAR-CD v4.06. Same configuration used for all both benchmark results - A-Class and C-Class.

• •

• •

Source: Published/submitted/approved results March 30, 2009.

101

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms HPC - Schlumberger Eclipse*, CMG* IMEX* and Landmark Nexus*



Reservoir Simulation using Schlumberger Eclipse*

• • •

Performance comparison based on results approved by ISV as of March 30, 2009. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM), 64-bit RedHat Enterprise Linux 5.3* (64-bit) and Eclipse version 2008.1 software. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 24GB (12x2GB 1066MHz DDR3) memory, 64-bit RedHat Enterprise Linux 5.3. Eclipse version 2008.1 software.

• •



Reservoir Simulation using CMG* IMEX*

• • •

Performance comparison based on results approved by ISV as of March 30, 2009. Intel® Xeon® processor 5160 based platform details Dell Precision 690 platform* with two Intel Xeon processors 5160 3.0GHz, 4MB L2 cache, 1333MHz FSB, 32GB memory, Microsoft Windows XP* 64bit OS. CMG IMEX, Version 2008.11. Intel® Xeon® processor X5482 based platform details Dell Precision T7400 platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB RedHat Enterprise Linux 5.2*. (64-bit) CMG IMEX, Version 2008.11. Intel® Xeon® processor X5570 based platform details Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 18GB memory. RedHat Enterprise Linux 5.3. (64-bit) CMG IMEX, Version 2008.11.

• • • •



Reservoir Simulation using Landmark Nexus*

• • •

Performance comparison based on results approved by ISV as of March 30, 2009. Intel® Xeon® processor X5482 based platform details Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM), 64-bit RedHat Enterprise Linux 5.2*. (64-bit) . Landmark Nexus R5000 software*. Intel® Xeon® processor X5560 based platform details Supermicro X8DTN+ server platform* with two Intel Xeon processors X5560 2.80GHz, 8MB L3 cache, QPI 6.4 MT/sec, 12GB memory, 64-bit RedHat Enterprise Linux 5.3. Landmark Nexus R5000 software.

• •

Source: Published/submitted/approved results March 30, 2009.

102

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

HPC cluster – ANSYS* FLUENT*, LS-DYNA*, SPEC MPI* 2007, MM5 v4.7.4, WRF v3.0.1 •

Computational Fluid Dynamics analysis using ANSYS* FLUENT (Cluster aircraft_2m and sedan_4m)



Comparison based on published/submitted results to www.fluent.com/software/fluent/fl6bench/fl6bench_6.4.x/index.htm as of March 30, 2009. Comparison uses cluster of size 1, 2, 4 and 8 nodes with 8 cores per node. Intel® Xeon® processor X5482 based platform details Cluster built with Supermicro* X7DB8+ server platforms each with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM), 64-bit RedHat Enterprise Linux 5.3* and Infiniband* interconnect. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.13)*. Intel® Xeon® processor X5560 based platform details Cluster built with Supermicro X8DTN+ server platforms each with two Intel Xeon processors X5560 2.80GHz, 8MB L3 cache, QPI 6.4 MT/sec, 18GB memory (6x2GB and 6x1GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3 and Infiniband interconnect. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.13). Quad-Core AMD Opteron* processor model 2384 platform based details Server platform with two AMD Opteron 2384 processor 2.7GHz, 6MB L3 cache, Linux OS. Cluster results published at www.fluent.com/software/fluent/fl6bench/fl6bench_6.4.x/index.htm.

• • • • • •



Crash Simulation analysis using LS-DYNA* (Cluster – 3 Vehicle Collision, Car2Car)

• • •

Comparison based on published/submitted results to http://www.topcrunch.org/ as of March 30, 2009. Comparison uses cluster of size 1, 2, 4, 8 and 16 nodes with 8 cores per node. Intel® Xeon® processor X5482 based platform details Cluster built with Supermicro X7DB8+* server platforms each with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM), 64 64-bit bit RedHat Enterprise Linux Linux* 5.3. Infiniband Infiniband* interconnect. Performance measured using LS-DYNA LS DYNA mpp971. Intel® Xeon® processor X5570 based platform details Cluster built with SGI Altix ICE 8200EX* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack* 6SP2. Performance measured using LS-DYNA mpp971.

• •



Performance of compute intensive applications using MPI with SPEC MPI* 2007



Comparison based on published at http://www.spec.org/mpi2007/results/mpi2007.html as of March 30, 2009. Comparison uses cluster of size 2, 4 , 8, 16 and 32 nodes with 8 cores per node MPI ranks equals number of compute cores enabled node,. enabled. Comparison uses power of two MPI rank counts from 16 to 128 in fully-subscribed fully subscribed mode on each cluster node Intel® Xeon® processor E5472 based platform details Cluster using SGI Altix ICE 8200EX server platforms* each with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FBDIMM), 64-bit SUSE LINUX 10* SP1 (X86-64) Intel® Xeon® processor X5570 based platform details Cluster built with SGI Altix ICE 8200EX server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI* 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server 10* SP2 with ProPack* 6SP2.

• • • •



Weather Simulation using MM5 v4.7.4 - t3a and WRF v3.0.1 - 12km CONUS :

• • •

Comparison based on measured results as of March 30, 2009. Comparison uses cluster of size 1, 2, 4, 8 and 16 nodes with 8 cores per node. Same platform used for both benchmarks Baseline Intel® Xeon® processor X5482 based platform details: Cluster built with SGI Altix ICE 8200EX* server platform each with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*. Intel® Xeon® processor X5570 based platform details: Cluster built with SGI Altix ICE 8200EX* server platform each with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*..

• •

Source: Published/submitted/approved results March 30, 2009.

103

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms 1P to 2P Scaling – Benefits of full configuration



1P to 2P scaling g for Intel® Xeon® processor p X5570



All measurements done using the same server platform. SuperMicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz). Measurement done using one and two processor configuration.

• •

Benchmark specific details SPECint rate base2006* and SPECfp_rate_base2006*: SPECint_rate_base2006* SPECfp rate base2006*: SUSE Linux Enterprise Server* 10 SP2 x86_64 x86 64 Kernel* 2 2.6.16.60-0.34-smp, 6 16 60 0 34 smp Intel® C++ Compiler for Linux32* and Linux64* version 11.0 build 20010131. LINPACK: RedHat Enterprise Linux 5.3* for x86_64 Kernel-2.6.18-128, Intel® SMP LINPACK* 10.2 beta version. SPECjbb2005*: Microsoft Windows Server 2008 Enterprise x64 Edition*, Oracle JRockit* 6 P28.0.0 (build P28.0.0-14-111048-1.6.0_0520090303-1104-windows-x86_64).

• •

Source: Published/submitted/approved results March 30, 2009.

104

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms SPEC CPU2006 suite on Xeon 5500 series – All bins

• • • • •

Xeon 5500 Series SPEC CPU2006 suite All results below are published by Intel Corporation at www.spec.org as of March 31, 2009 All data measured with Intel Compiler 11.0, Feb 2009 binaries, SLES 10 SP 2 OS (kernel 2.6.16.60-0.34-smp). D-0 stepping processor. Note that we don’t actually have 800 MHz memory. It is the same 1066 MHz memory but it would run at 800Mhz when used with the lower skus. SPEED

intpeak p

Rate

Memory

Sku details

Part #

IntBase

Nehalem-EP 1.86/4.80 GT/s (DC)

E5502

17.9

19.9

HTML

21.9

23.1

HTML

66.2

71.3

HTML

68.1

70.9

HTML

24GB (12x2GB) DDR3 800

Nehalem-EP 2.00/4.80 GT/s

E5504

19.2

21.3

HTML

23.7

25.2

HTML

127

136

HTML

111

115

HTML

24GB (12x2GB) DDR3 800

Nehalem-EP 2.13/4.80 GT/s

E5506

20.2

22.5

HTML

24.6

26.3

HTML

133

142

HTML

115

119

HTML

24GB (12x2GB) DDR3 800

Nehalem-EP 2.13/4.80 GT/s (LV)

L5506

20.3

22.5

HTML

24.5

26.5

HTML

133

142

HTML

115

119

HTML

24GB (12x2GB) DDR3 800

Nehalem-EP 2.26/5.86 GT/s

E5520

24.7

27.8

HTML

30.1

32.0

HTML

187

202

HTML

154

160

HTML

24GB (12x2GB) DDR3 1066

Nehalem-EP 2.26/5.86 GT/s (LV)

L5520

24.7

27.9

HTML

30.4

32.2

HTML

183

198

HTML

152

158

HTML

24GB (12x2GB) DDR3 1066

Nehalem-EP 2.40/5.86 GT/s

E5530

25.8

29.1

HTML

31.3

33.2

HTML

194

210

HTML

159

165

HTML

24GB (12x2GB) DDR3 1066

Nehalem-EP 2.53/5.86 GT/s

E5540

27.0

30.4

HTML

32.4

34.3

HTML

201

217

HTML

162

169

HTML

24GB (12x2GB) DDR3 1066

Nehalem-EP 2.66/6.40 GT/s

X5550

29.7

33.5

HTML

35.8

38.2

HTML

224

241

HTML

183

189

HTML

24GB (6x4GB) DDR3 1333

Nehalem-EP 2.80/6.40 GT/s

X5560

30.8

34.6

HTML

37.1

39.4

HTML

231

248

HTML

187

194

HTML

24GB (6x4GB) DDR3 1333

Nehalem-EP 2.93/6.40 GT/s

X5570

31.9

35.9

HTML

38.3

40.8

HTML

237

255

HTML

191

198

HTML

24GB (6x4GB) DDR3 1333

fpbase

Fppeak p

Intratebase

Intratepeak p

Fpratebase

Fpratepeak p

Memory

HT ON; Turbo ON; (NA for 5502-06) D-0 Step; IC 11.0 binaries (Feb 09) SLES 10 SP2(kernel 2.6.16.60-0.34-smp)

Source: Published/submitted/approved results March 30, 2009.

105

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer Throughput performance on SPECint*_rate_base2006

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

Irwindale 3.8 http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00027.html Xeon 5110 ((WDC 1.60)) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02272.html p p g p q p Xeon 5120 (WDC 1.86) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02276.html Xeon 5130 (WDC 2.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071112-02568.html Xeon 5140 (WDC 2.33) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02185.html Xeon 5150 (WDC 2.66) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071112-02565.html Xeon 5160 (WDC 3.0) - http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070803-01638.html Xeon 5405 ((HTN 2.0)) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081208-06236.html p p g p q p Xeon 5410 (HTN 2.33) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081124-06135.html Xeon 5420 (HTN 2.50) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05755.html Xeon 5430 (HTN 2.66) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081031-05828.html Xeon 5440 (HTN 2.83) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05610.html Xeon 5450 (HTN 3.0) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081109-05863.html Xeon eo 5 5460 60 ((HTN 3 3.16) 6) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05750.html ttp // spec o g/cpu 006/ esu ts/ es 008q /cpu 006 008 0 05 50 t Xeon 5470 (HTN 3.33) -http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080901-05156.html Xeon 5462 (HTN 2.8/1600) – Estimate based on Intel internal measurement – March 2009 Xeon 5472 (HTN 3.0/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081030-05826.html Xeon 5482 (HTN 3.2/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05590.html Opteron 2347 (Bar 1.9) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080811-05001.html Opteron 2350 (Bar 2.0) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006 http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080414-04112.html 20080414 04112.html Opteron 2352 (Bar 2.1) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080414-04114.html Opteron 2354 (Bar 2.2) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080331-04011.html Opteron 2356 (Bar 2.3) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080623-04661.html Opteron 2358 (Bar 2.4) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080801-04849.html Opteron 2360 (Bar 2.5) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080801-04850.html Opteron 2376 (Sh 2 2.3) 3) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006 http://www spec org/cpu2006/results/res2009q1/cpu2006-20081218-06300 20081218 06300.html html Opteron 2378 (Sh 2.4) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090105-06386.html Opteron 2380 (Sh 2.5) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081208-06244.html Opteron 2382 (Sh 2.6) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06321.html Opteron 2384 (Sh 2.7) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081024-05683.html Opteron 2386 (Sh 2.8) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06302.html Opteron 2389 (Sh 2 2.9) 9) - http://www.spec.org/cpu2006/results/res2009q2/cpu2006 http://www spec org/cpu2006/results/res2009q2/cpu2006-20090216-06529 20090216 06529.html html

Source: Published/submitted/approved results March 30, 2009.

106

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating-Point Throughput performance on SPECfp*_rate_base2006

• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

Irwindale 3.8 http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00025.html Xeon 5110 (WDC 1.60) 1 60) - http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070403-00762.html http://www spec org/cpu2006/results/res2007q2/cpu2006 20070403 00762 html Xeon 5120 (WDC 1.86) - http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070403-00770.html Xeon 5130 (WDC 2.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02205.html Xeon 5140 (WDC 2.33) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02188.html Xeon 5150 (WDC 2.66) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02035.html Xeon 5160 (WDC 3.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02036.html X Xeon 5405 (HTN 2 2.0) 0) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05770.html htt // / 2006/ lt / 2008 4/ 2006 20081027 05770 ht l Xeon 5410 (HTN 2.33) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081124-06029.html Xeon 5420 (HTN 2.50) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05462.html Xeon 5430 (HTN 2.66) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081110-05913.html Xeon 5440 (HTN 2.83) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090202-06483.html Xeon 5450 (HTN 3.0) -http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081216-06289.html X Xeon 5460 (HTN 3 3.16) 16) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05748.html htt // / 2006/ lt / 2008 4/ 2006 20081027 05748 ht l Xeon 5470 (HTN 3.33) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080908-05235.html Xeon 5462 (HTN 2.8/1600) - Estimate based on Intel internal measurement – March 2009 Xeon 5472 (HTN 3.0/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081022-05675.html Xeon 5482 (HTN 3.2/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05587.html Opteron 2347 (Bar 1.9) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080811-05002.html Opteron 2350 (Bar 2.0) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080414-04111.html Opteron 2352 (Bar 2.1) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080414-04113.html Opteron 2354 (Bar 2.2) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080414-04115.html Opteron 2356 (Bar 2.3) - http://www.spec.org/cpu2006/results/res2008q2/cpu2006-20080317-03774.html Opteron 2358 (Bar 2.4) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080801-04843.html Opteron 2360 (Bar 2.5) - http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080801-04851.html Opteron 2376 (Sh 2.3) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090119-06451.html Opteron 2378 (Sh 2.4) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090105-06385.html Opteron 2380 (Sh 2.5) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06317.html Opteron 2382 (Sh 2.6) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06320.html Opteron 2384 (Sh 2.7) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090116-06418.html Opteron 2386 (Sh 2.8) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06303.html

Source: Published/submitted/approved results March 30, 2009.

107

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Integer “SPEED” performance on SPECint*_base2006

• • • • • • • • • • • • • • • • • • • • • • • •

Irwindale 3.8 http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00028.html Xeon 5110 (WDC 1.60) 1 60) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02273.html http://www spec org/cpu2006/results/res2007q4/cpu2006 20071015 02273 html Xeon 5120 (WDC 1.86) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02277.html Xeon 5130 (WDC 2.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02203.html Xeon 5140 (WDC 2.33) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02186.html Xeon 5150 (WDC 2.66) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02043.html Xeon 5160 (WDC 3.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02044.html X Xeon 5405 (HTN 2 2.0) 0) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081109-05870.html htt // / 2006/ lt / 2008 4/ 2006 20081109 05870 ht l Xeon 5410 (HTN 2.33) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05785.html Xeon 5420 (HTN 2.50) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05757.html Xeon 5430 (HTN 2.66) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081124-06046.html Xeon 5440 (HTN 2.83) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081110-05962.html Xeon 5450 (HTN 3.0) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081208-06207.html X Xeon 5460 (HTN 3 3.16) 16) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05771.html htt // / 2006/ lt / 2008 4/ 2006 20081027 05771 ht l Xeon 5470 (HTN 3.33) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05450.html Xeon 5462 (HTN 2.8/1600) Xeon 5472 (HTN 3.0/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081030-05825.html Xeon 5482 (HTN 3.2/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05592.html Opteron 2376 (Sh 2.3) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090119-06454.html Opteron 2378 (Sh 2.4) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090105-06387.html Opteron 2380 (Sh 2.5) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06319.html Opteron 2382 (Sh 2.6) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06323.html Opteron 2384 (Sh 2.7) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081110-05930.html Opteron 2386 (Sh 2.8) -

Source: Published/submitted/approved results March 30, 2009.

108

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Floating Point “SPEED” performance on SPECfp*_base2006

• • • • • • • • • • • • • • • • • • • • • • • •

Irwindale 3.8 http://www.spec.org/cpu2006/results/res2006q3/cpu2006-20060513-00026.html Xeon 5110 (WDC 1.60) 1 60) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02282.html http://www spec org/cpu2006/results/res2007q4/cpu2006 20071015 02282 html Xeon 5120 (WDC 1.86) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071015-02280.html Xeon 5130 (WDC 2.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02198.html Xeon 5140 (WDC 2.33) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20071001-02189.html Xeon 5150 (WDC 2.66) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02037.html Xeon 5160 (WDC 3.0) - http://www.spec.org/cpu2006/results/res2007q4/cpu2006-20070916-02038.html X Xeon 5405 (HTN 2 2.0) 0) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081215-06287.html htt // / 2006/ lt / 2009 1/ 2006 20081215 06287 ht l Xeon 5410 (HTN 2.33) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081124-06139.html Xeon 5420 (HTN 2.50) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05754.html Xeon 5430 (HTN 2.66) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081124-06044.html Xeon 5440 (HTN 2.83) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05414.html Xeon 5450 (HTN 3.0) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081109-05873.html X Xeon 5460 (HTN 3 3.16) 16) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081027-05767.html htt // / 2006/ lt / 2008 4/ 2006 20081027 05767 ht l Xeon 5470 (HTN 3.33) -http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080915-05337.html Xeon 5462 (HTN 2.8/1600) Xeon 5472 (HTN 3.0/1600) - http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081030-05824.html Xeon 5482 (HTN 3.2/1600) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081215-06283.html Opteron 2376 (Sh 2.3) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090119-06453.html Opteron 2378 (Sh 2.4) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090105-06384.html Opteron 2380 (Sh 2.5) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06316.html Opteron 2382 (Sh 2.6) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081222-06322.html Opteron 2384 (Sh 2.7) - http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090116-06416.html Opteron 2386 (Sh 2.8) -

Source: Published/submitted/approved results March 30, 2009.

109

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Java performance on SPECjbb*2005



SPECjbb2005: j Configuration g Details: Based on Intel Internal measured data and p published results. March 2009



Intel Xeon Processor with 2M L2 Cache based platform details: Intel Server pre-production platform with two 64-bit Intel® Xeon™ processors with 2MB L2 Cache and 800 MHz system bus 8GB (8x1GB) DDR2 400MHz memory; Microsoft Windows Server 2003 Standard Edition. BEA JRockit(R) 5.0 JVM. Run with 1 JVM instance. Same configuration used for the different frequencies. Dual-Core Intel Xeon Processor 51xx based p platform details: Intel Server p pre-production p hardware with two DualCore Intel Xeon Processor 5100 series, 8GB (8x1GB) 667MHz FB-DIMM memory, Microsoft Windows Server* 2003. BEA JRockit(R) 5.0 P27. Run with one jvm instances. Same configuration used for the different frequencies. Quad-Core Intel Xeon Processor 54xx based platform details: Intel Server pre-production platform with two QuadCore Intel Xeon Processor 5400 series, 16GB (8x2GB) 667MHz FB-DIMM memory, Microsoft Windows Server* 2003. BEA JRockit(R) 5.0 P27. Run with four jvm instances. Same configuration used for the different frequencies. Quad-Core Intel Xeon Processor 55xx based platform details: Intel Server pre-production platform with two QuadC Core IIntel t lX Xeon P Processor 5500 series, i 12GB (6 (6x2GB) 2GB) 1333/1066MH 1333/1066MHz DDR3 memory, Microsoft Mi ft Windows Wi d S Server** 2008 SP1. Oracle JRockit(R) 5.0 P28. Run with four jvm instances.

• • •

• • • •

Opteron 2356: http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080506-00487.html Opteron 2360: http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090224-00587.html Opteron 2376: http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20081218-00566.html Opteron 2386: http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20081218-00567.html

Source: Published/submitted/approved results March 30, 2009.

110

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Energy Efficienct performance on SPECpower*_ssj 2008



SPECpower_ssj 2008 Configuration Details: Based on Intel Internal measured data results. March 2009



Quad-Core Intel Xeon Processor 54xx based platform details: Intel Server pre-production platform with two QuadCore Intel Xeon Processor 5400 series, 8GB (4x 2GB) 667MHz FB-DIMM memory, Microsoft Windows Server* 2003 SP2. BEA JRockit(R) 5.0 P27.5. Run with 2 jvm instances. Quad-Core Intel Xeon Processor 55xx based platform details: Intel Server pre-production platform with two QuadCore Intel Xeon Processor 5500 series, 8GB (4x 2GB) 1333/1066MHz DDR3 memory, Microsoft Windows Server* 2008 SP1 Oracle SP1. O l JRockit(R) JR kit(R) 5 5.0 0 P28 P28.0. 0 Run R with ith 2 jjvm iinstances t



Source: Published/submitted/approved results March 30, 2009.

111

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms Dense Floating-Point performance on LINPACK

• Intel® Xeon® Processor 3.60 GHz based platform details: Intel® internal measurement. Intel® preproduction software development platform with two 64-bit Intel® Xeon® processors 3.60 GHz with 2MB L2 Cache and 800 MHz system bus and 8GB (8x1024 MB) DDR2-400 memory; Red Hat Enterprise Linux 3, Update 3, EM64T, Intel® Linpack 2.1.2, Problem Size = 31,000. processor 5160 based p platform details: Supermicro* p X7DB8+ server • Dual-Core Intel® Xeon® p platform with two Dual-Core Intel Xeon processors 5160 3.00GHz, 4MB L2 cache, 1333MHZ FSB, 16GB memory (8x2GB FBD 667MHz), 64-bit RedHat Enterprise* Linux* Server 5. Intel® SMP Linpack Version 10.0, Problem Size=44,000. • Quad-Core Intel® Xeon® processor 54xx (Bensley) based platform details: Supermicro X7DB8+ server platform with two Quad-Core Intel Xeon processors 5400 series, 2x6MB L2 cache, 1333MHZ FSB, 16GB memory (8x2GB FBD 667MHz), 64-bit RedHat Enterprise* Linux* Server 5 Update 2. 2 Intel® SMP Linpack Version 10 10.1, 1 Problem Size=45,000. Size=45 000 • Quad-Core Intel® Xeon® processor 54xx based (Stoakley) platform details: Supermicro X7DWN+ server platform with two Quad-Core Intel Xeon processors 5400 series, 2x6MB L2 cache, 1600MHZ FSB, 16GB memory (8x2GB FBD 800MHz), 64-bit RedHat Enterprise* Linux* Server 5 Update 3. Intel® SMP Linpack Version 10.2 Beta, Problem Size= 45,000. • Quad-Core Intel® Xeon® processor 55xx based platform details: Supermicro X8DTN+ server platform with two Quad Quad-Core Core Intel Xeon processors 5500 series series, 8M L3 cache cache, 24GB memory (12x2GB DDR3 1333/1066MHz), 64-bit RedHat Enterprise* Linux* Server 5 Update 3. Intel® SMP Linpack Version 10.2 Beta and 10.2, Problem Size=55,000.

Source: Published/submitted/approved results March 30, 2009.

112

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Intel® Xeon® Processor 5500 series based Server platforms

Configuration for Server Consolidation & Comparison with selected older generation platform Year

Code Name

Brand Name

2001

Tualatin

2002

Prestonia

Launch

Xeon Freq

Intel Pentium III 1.40Ghz 512M L2 WW32 Intel Xeon 2.20 GHz 512M L2

Cores/ CPU

L1 Cache

L2 Cache

L3 Cache

FSB

Chipset

SPECint_ SPECint_ SPECfp_ SPECfp_ rate*_ rate*_ rate*_ rate*_ base2000 base2006 base2000 base2006

TPC-C* tpmC

SAPSD* Users

Linpack SPECjbb GFlops 2005

1.26

1

16K(I)+16K(D) 512KB

0

133

SW HE-SL

12.2

6.5

17079

135

-

April

2.20

1

12K(I)+8K(D) 512KB

0

400

E7500

15.7

13.1

33768

215

5.46

2002

Prestonia

Intel Xeon 2.80 GHz 512M L2

Sept

2.80

1

12K(I)+8K(D)

512KB

0

400

E7500

18.9

14.4

39007

245

6.72

2003

Prestonia

Intel Xeon 3.06 GHz 512M L2

March

3.06

1

12K(I)+8K(D)

512KB

0

533

E7501

21.5

16.7

44943

290

7.83

2003

Prestonia

Intel Xeon 3.06 GHz 1M L3

July

3.06

1

12K(I)+8K(D)

512KB

1MB

533

E7501

26.0

20.4

52468

329

8.16 8.40

2003

Prestonia

Intel Xeon 3.20 GHz 1M L3

Oct

3.20

1

12K(I)+8K(D)

512KB

1MB

533

E7501

26.9

20.7

54097

340

2004

Prestonia

Intel Xeon 3.20 GHz 2M L3

March

3.20

1

12K(I)+8K(D)

512KB

2MB

533

E7501

30.7

24.0

60364

408

9.79

2004

Nocona

Intel Xeon 3.60 GHz 1M L2

June

3.60

1

16K(I)+16K(D)

1MB

0

800

E7520

32.1

25.2

68010

470

11.60

2005

Irwindale

64-bit Intel Xeon 3.60 GHz 2M L2

Feb

3.60

1

12K(I)+16K(D)

2MB

0

800

E7520

39.9

74298

597

13.00

2006

Irwindale

64-bit Intel Xeon 3.80 GHz 2M L2

Sept

3.80

1

12K(I)+16K(D)

2MB

0

800

E7520

43.3

2005 Paxville DP Dual-Core Intel Xeon 2.80 2x2M L2

Oct

2.80

2

12K(I)+16K(D) 2x2MB

0

800

E7520

59.5

87791

788

18.10

125954

1047

22.10

169360

1285

41.70

240737

1870

73.00

251300

2080

77.10 81.67

2006

31.6 20.9

33.1

18.8

39.4

Dempsey

Dual-Core Dual Core Intel Xeon 5080

May

3.73

2

12K(I)+16K(D) 12K(I) 16K(D) 2x2MB

0

1066

5000p

81.2

2006 Woodcrest

Dual-Core Intel Xeon 5160

June

3.00

2

32K(I)+32K(D)

0

1333

5000p

123.0

2006 Clovertown

Quad-Core Intel Xeon X5355

Nov

2.66

4

32K(I)+32K(D) 2x4MB

0

1333

5000p

200.0

2007 Clovertown

Quad-Core Intel Xeon X5365

July

3.00

4

32K(I)+32K(D) 2x4MB

0

1333

5000p

214

2007 Harpertown

Quad-Core Intel Xeon X5460

Nov

3.16

4

32K(I)+32K(D) 2x6MB

0

1333

5000p

242

131

275149

2449

2007 Harpertown

Quad-Core Intel Xeon E5462

Nov

3.00

4

32K(I)+32K(D) 2x6MB

0

1600

5400

252

149

-

-

80.6

2008 Harpertown

Quad Core Intel Xeon X5470 Quad-Core

July

3 33 3.33

4

32K(I)+32K(D) 2x6MB

0

1333

5000p

-

2518

81 67 81.67

2008 Harpertown

Quad-Core Intel Xeon X5482

July

3.20

4

32K(I)+32K(D) 2x6MB

0

1600

80.6

2009 Nehalem-EP

Intel Xeon Processor X5570

Mar

2.93

4

32K(I)+32K(D) 4x256KB 8MB 6.4QPI

Intel Pentium® III Microarchitecture Intel Penryn Microarchitecture



Note – – –

4MB

61.6

50970

60.8

82.1

45.1

103.0 99.1

105

63.1

140

5400

147

86.4

-

-

5500

241

194

631766

5100

Intel NetBurst® Microarchitecture

138388 252403

368034 604417

Intel Core® Microarchitecture

Intel Nehalem Microarchitecture

Only selected data used in the charts; Additional data provided in the table above for reference Information in Bold indicates major changes from one generation to other Intel launch dates and details at http://www.intel.com/pressroom/kits/quickreffam.htm

Source: Published/submitted/approved results March 30, 2009.

113

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Nehalem-EP Platform Comparison Bandwidth-Latency Curve

Bandwidth-Latency Curve – DP platforms 400

HTN/BF1333 - 75% Read

Read Latency (ns)

Latency: Low wer is better

350

HTN/SB1600 - 75% Read

NHM 2.93 - 75% Read

300 250 200 150 100 50 0 0

5000

10000

15000

20000

25000

30000

35000

40000

Bandwidth (MBs/Sec) Bandwidth: Higher is better

Significant improvement in bandwidth and latency Source: Intel Internal measurements Nov 2008. See backup for additional details Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or 114 design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on software performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.

Suggest Documents