Intelligent Mainframe Management with CA

For Informational Purposes Only Terms of this Presentation © 2016 CA. All rights reserved. All trademarks referenced herein belong to their respective companies. The presentation provided at New England DB2 Users Group is intended for information purposes only and does not form any type of warranty. Some of the specific slides with customer references relate to customer's specific use and experience of CA products and solutions so actual results may vary. Certain information in this presentation may outline CA’s general product direction. This presentation shall not serve to (i) affect the rights and/or obligations of CA or its licensees under any existing or future license agreement or services agreement relating to any CA software product; or (ii) amend any product documentation or specifications for any CA software product. This presentation is based on current information and resource allocations as of September 1, 2016, and is subject to change or withdrawal by CA at any time without notice. The development, release and timing of any features or functionality described in this presentation remain at CA’s sole discretion. Notwithstanding anything in this presentation to the contrary, upon the general availability of any future CA product release referenced in this presentation, CA may make such release available to new licensees in the form of a regularly scheduled major product release. Such release may be made available to licensees of the product who are active subscribers to CA maintenance and support, on a when and ifavailable basis. The information in this presentation is not deemed to be incorporated into any contract.

Agenda ROLE OF MAINFRAME IN THE MODERN APP ECONOMY WHAT ARE OUR CUSTOMERS PRIORITIES? WHAT IS CA DOING TO ADDRESS THOSE PRIORITIES PREVIEW NEXT VERSION OF SYSTEMS MANAGEMENT REVIEW DESIGN THINKING PROCESS, USE CASES DEMO AND DEEP DIVE

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Mainframe in the modern app economy

70% of

ANOMALY DETECTION

EXPERT SYSTEMS

64% increase in mainframe workloads

transactions flow through mainframe

55%

70%

of apps depend on mainframe

world’s corporate data is on a mainframe

BUSINESS SERVICE MANAGEMENT © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

SECURITY BREACH DETECTION

Our Goal - Building the Intelligent Mainframe USER PERSONA

BI/Statistical Modeling

Automation

SECURITY BREACH DETECTION

Automation ENTERPRISE SUPPORT

Data Driven Operations

Data Analytics

GENERALIST TOOLS Generalist

SPECIALIST TOOLS

Discover & monitor business service topology

Root cause hypothesis

Focused resolution guidance

Automate event response

EXPERT SYSTEMS

Predict business service performance

BUSINESS SERVICE MANAGEMENT

MAINFRAME OPERATIONS

Specialists

REPORTING

Experts

ANOMALY DETECTION Predict and detect anomalies, respond to anomaly alerts

EXPERT TOOLS Monitor Status, react to threshold based events

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Cost

Progress To-date – Where does the transition start? BENEFITS

SOLUTIONS  Database Management – Expands value of CA Detector – Currently in Validation



Makes existing solutions more intelligent



Focus on problem avoidance via anomaly detection & dynamic thresholds



Simplified U/X designed for quick and easy collaboration

 Performance Management – Expands value of CA SYSVIEW – Validation in June – GA ~ Fall 2016 Register to participate in validation at:

validate.ca.com © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Use Cases Problem, Opportunity and Proposed Solution

Mainframe IT Ops Team Persona PETE

RALPH

Level 1 Support Analyst Cross Enterprise

Systems Performance Analyst

SHERMAN

DEBBIE

FRED

Systems Engineer

Network Engineer

Application DBA

MY PAIN

MY PAIN

MY PAIN

Monitoring many systems & Devices

Bottleneck – work on all issues

Firefighting and identifying likely sources of future fires

HELP ME

HELP ME

HELP ME

Simplify alerts, meaning and action

Understand app and system performance characteristics quickly

Detect anomalies before they become an issue and collaboration to isolate the functional area

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Anomaly Detection – Problem Avoidance Use Case & Solution Concept

PERSONA:

• Systems Programmer, • Network Engineer, • Application DBA

USE CASE: Sherman, Debbie, and Fred have deep experience and skills in their respective areas. They hate being called after a system problem has occurred and being asked to prove that their area was at fault or innocent. They would like a system that tells them when their area is behaving abnormally, so that they can address it before it comes to the attention of the Systems Performance Engineer.

PROBLEM  “Avoid, detect and predict issues that might be a problem”  “Thresholds are hard to maintain and generate large amount of false positives”  “Ability to see multiple views of data ”  “Ability to create views of information faster”  “The current U/X prevents easy collaboration and access to analytics” © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Modern tools & views – Problem Avoidance/Remediation Use Case & Solution Concept

SOLUTION  Dashboards on demand and rapid actions  Leverage UX that complements SME existing workflow  Simplified U/X – browser access & designed for collaboration

BOTTOM LINE  High Availability  Problem avoidance  Reduced MTTR  Reduce SME dependence for issue detection © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Anomaly Detection– Problem Avoidance/Prediction Use Case & Solution Concept

SOLUTION  Detect anomalies & predict issues realtime, alert based on predefined rules  Leverage historical data and machine learning for dynamic thresholds  Simplified U/X – browser access & designed for collaboration

BOTTOM LINE  High Availability  Problem avoidance  Reduced MTTR  Reduce SME dependence for issue detection

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Magic Behind the Analytics – Data Science and Prediction  Goals:

Kernel Density Estimator

Historical Input Data, Same Stage of Business Cycle

– Dynamically and automatically determine baselines and thresholds Wavelet Decomposition

Predictive Model

ExponentiallyWeighted MovingAverage

– Generate alerts for abnormal scenarios, eliminate false positives and minimize false negatives  Business cycles  Natural volatility

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Solution Demo

Today’s Challenge 



Companies are developing complex SQL applications on DB2 –

Time is over where we (the DBA’s) knew the SQL transactions – dynamic SQL taking is over !

–

Performance often degrades over time without anyone noticing (the creeping trend)

–

The degradation can happen slowly (over weeks or months) or quickly (over hours or days)

When performance degrades, DBAs have steps they can take to “restore” performance –



But with thousands to hundreds of thousands of SQL statements executing, a DBA often does not recognize degradation until customers complain or service level agreements have been missed

DBAs need an early warning system – Predictive Reactive –

To recognize and prioritize where and when significant changes in SQL performance are occurring, before customers complain or service level agreements are missed

–

Intelligent dynamic baseline metrics to avoid false positives / too many false negatives.

–

The system needs to be intelligent, the interface needs to be intuitive, and customers cannot afford additional overhead on their z/OS systems © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Today - DBA struggles to identify what to tune To minimize elapsed time, CPU, and Getpages - while maintaining SLAs Methods

Adjusts SQL statements, runs reorg utility, tunes databases, etc.

Challenge

Finding best applications to tune – Detector really helps !!

Data

CA Detector interval performance metrics for executions of SQL statements, packages, and plans

Limitations CA Detector does a great job telling you where you are • • • •

How you got here (is today’s behavior “normal”?) Where you are going (is performance getting better or worse?) Where the greatest tuning potential is (which applications account for the greatest increases in resource usage ? ) Does is matter if a transaction is using 15 GETP as opposed to 10 one month ago ? Maybe – what if it executes 20M times/daily, and can you “normalize dynamic statements ? © 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Do you proactively monitor application performance or do you rely on user complaints? – Do you monitor trends to identify potential problems before they occur? – Do you manually determine, set, and adjust thresholds for monitoring applications (aka. Baselines)? – Does your current solution automatically determine what is “normal”? – Do you spend time reviewing false or short-term “spike” alerts / event notifications (as opposed to sustained deviations) ? – How much time is spent defining /maintaining thresholds?

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

How do you resolve performance problems? 

How do you determine when performance started to degrade?



How long does it take to identify the problematic SQL statements?



Do you know how the statements executed before the performance degraded?



Do you have a log of application performance problems and resolutions? – Can you compare this log with current problems?



Many customers offload Detector datastore into DB2 tables –

Query heaviest plans/packages & manually compare to “baselines”

–

No efficient method to re-evaluate baselines when “the world changes”

–

How to monitor “standard deviation” and “creeping trend” is complex and cumbersome

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution

Thank you

Jim Endler Sr. Principal Consultant, Technical Sales CA Technologies | 11325 N. Community House Road Suite 550 | Charlotte, NC 28277 Mobile: +1 713 703 5888 | [email protected]

© 2016 CA. All rights reserved. CA confidential and proprietary information. No unauthorized use, copying or distribution