Copyrighted Material for Preview Use Only

Introduction to TCP/IP Unit 1 TCP/IP Structure and Addressing This unit introduces the Transmission Control Protocol/Internet Protocol (TCP/IP) protocol suite. We begin with a review of the Open Systems Interconnection (OSI) model. We then introduce the TCP/IP Network, Transport, and Application Layer protocols. Then we explore the different address types necessary to move information from one user’s application to another’s across a TCP/IP network. We review IP address structures and how they are divided into logical and physical groups called subnets. To end the unit, we review some of the most common network devices responsible for moving data across networks.

Lessons 1. OSI Model Review 2. Internet Protocol Suite 3. Internet Addressing 4. Internetworking Devices

Terms 100BaseFX—100BaseFX is a 100BaseT variant that runs over fiber optic cabling. 100BaseFX is generally used for high-speed LAN backbones.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

1

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP 100BaseT—100BaseT is based on 802.3 Ethernet and uses twisted pair cabling, as does 10BaseT Ethernet. However, 100BaseT runs 10 times faster than 10Base2 and 10Base5, at 100 Mbps. 10Base2—10Base2 is one of several physical cabling standards for 802.3 Ethernet networks. 10Base2 uses thin RG-58A/U coaxial cabling for connecting nodes on a bus topology, as does 10Base5. It is also known as Thinnet. 10Base5—Also known as Thicknet, 10Base5 was the original Ethernet cabling standard. It is called Thicknet because it uses a coaxial cable approximately 0.5 inches thick. 10BaseT—10BaseT is one of several physical cabling standards for 802.3 Ethernet LANs. 10BaseT networks use twisted pair cabling for connecting nodes in a star topology. Address Resolution Protocol (ARP)—ARP is the protocol used by IP (as in TCP/IP) for address resolution. Address resolution refers to the ability of a station to resolve another station’s MAC (hardware) address given its IP address. AppleTalk—AppleTalk is Apple’s proprietary, seven-layer, peer-topeer network communications protocol for Macintosh networks. AppleTalk runs on Apple network topologies, as well as over Ethernet and Token Ring networks. Asynchronous Transfer Mode (ATM)—ATM is a cell-switching network that consists of multiple ATM switches that forward each individual cell to its final destination. ATM can provide transport services for audio, data, and video. Bootstrap Protocol (BOOTP)—BOOTP, described in RFC 951, is a method in which a diskless workstation or terminal can obtain configuration information, such as its IP address and a configuration file, from a server on which this information is stored. bridge—A bridge is a hardware device that connects LANs. It can be used to connect LANs of the same type, such as two Token Ring segments, or LANs with different types of media, such as Ethernet and Token Ring. A bridge operates at the Data Link Layer of the OSI reference model. bridge router (brouter)—A brouter is an internetworking device that combines the functions of both a bridge and a router. See router.

2

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Overview broadcast—The term broadcast is used in several different ways in communications and networking. With respect to LANs, the term refers to information (that is, frames) sent to all devices on the physical segment. For example, a bus topology, in which a common cable is used to connect devices, is considered a broadcast technology. Another common use of the term broadcast relates to frames. Broadcast frames contain a special destination address that instructs all devices on the network to receive the frame. checksum—Checksum is computed by the number of bits in a transmitted unit of data. A checksum is appended to a data unit as a simple error-detection method. The receiving node counts the data bits and compares the result to the checksum, to see whether all bits have arrived. If the numbers match, the transmission was likely complete. TCP and UDP provide checksum count and verification. collision—A collision occurs in an Ethernet network when two frames are put onto the physical medium at the same time and overlap fully or partially. When a collision occurs, the data on the physical segment is no longer valid. datagram—A datagram is a unit of information processed by the Network Layer of the OSI reference model. The packet header contains the logical (network) address of the destination node. Intermediate nodes forward a packet until it reaches its destination. A packet can contain an entire message generated by higher OSI layers or a segment of a much larger message. DECnet—DECnet is a proprietary network architecture created by DEC (now Hewlett-Packard). The most recent iteration, DECnet Phase V, specifies seven layers that correspond to the OSI reference model, while maintaining compatibility with previous versions. Domain Name System or Service (DNS)—In a TCP/IP network, a user can communicate with another user by specifying a name, such as [email protected]. TCP and IP require Internet addresses for messages, thus one must be translated to the other. This is the job of the DNS; given a name, it returns an Internet address.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

3

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP Dynamic Host Configuration Protocol (DHCP)—DHCP provides configuration parameters to Internet hosts. It consists of two components: a protocol for delivering host-specific configuration parameters from a DHCP server to a host, and a mechanism for allocation of network addresses to hosts. DHCP is built on a client/ server model, where designated DHCP server hosts allocate network addresses and deliver configuration parameters to dynamically configured hosts. Ethernet—Ethernet technology, originally developed in the 1970s by Xerox Corporation in conjunction with Intel and DEC, is now the primary medium for LANs. The original Ethernet has 10-Mbps throughput and uses the CSMA/CD method to access the physical media. Fast Ethernet (100-Mbps Ethernet) and Gigabit Ethernet (1,000-Mbps Ethernet) are also used. expansion card—An expansion card is a device you plug in to a PC backplane to add functionality, such as a modem, NIC, or disk controller. extranet—An extranet is a broader form of a private intranet. Extranets are private TCP/IP networks that are shared between closely aligned organizations, and are not available to the general public. Fiber Distributed Data Interface (FDDI)—FDDI is a LAN standard specifying a 100-Mbps token-passing network using fiber optic cable. File Transfer Protocol (FTP)—FTP is a TCP/IP Application Layer protocol used to transfer files between two computers. hub—Also referred to as a wiring concentrator, a simple hub is a repeater with multiple ports. A signal coming into one port is repeated out the other ports. Hypertext Transfer Protocol (HTTP)—HTTP is the Application Layer protocol used to request and transmit documents by means of the Web. Institute of Electrical and Electronic Engineers (IEEE)—IEEE is a professional organization composed of engineers, scientists, and students. Founded in 1884, IEEE publishes computer and electronics standards, including the 802 series that defines shared-media networks such as Ethernet and Token Ring. Internet—The term “Internet,” capitalized, refers to the global internetwork of TCP/IP networks.

4

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Overview Internet Control Message Protocol (ICMP)—ICMP is an integral part of IP that handles error and control messages. Gateways and hosts use ICMP to report problems about datagrams back to the original source that sent the datagram. ICMP also includes an echo request/reply used to test whether a destination is reachable and responding. Internet Corporation of Assigned Names and Numbers (ICANN)— ICANN is a private, nonprofit organization responsible for overseeing the domain name registration process, assigning IP addresses, assigning protocol parameters, and managing the DNS root servers. Learn more about ICANN at http://www.icann.org. Internet Engineering Task Force (IETF)—IETF is a large, open, international community of network designers, operators, vendors, and researchers concerned with the evolution of the Internet architecture and smooth operation of the Internet. Internet Group Management Protocol (IGMP)—IGMP is the Internet standard by which hosts can communicate their multicast group membership status to multicast routers. IGMP is used to keep up-to-date information on which host is in which multicast group. Internet Packet Exchange/Sequenced Packet Exchange (IPX/ SPX)—IPX is NetWare’s proprietary Network Layer protocol. SPX is the connection-oriented transport protocol concerned with connection-oriented services such as sequencing packets and guaranteeing their delivery, which provides reliability for IPX communications. Internet Protocol (IP)—IP is the TCP/IP standard protocol that defines the IP datagram as the unit of information passed across an internet. IP provides the basis for connectionless, best-effort packet delivery service, and includes ICMP. The entire protocol suite is often referred to as TCP/IP because TCP and IP are the two most fundamental protocols. internetwork—A complex network that may combine smaller networks in different physical locations, based on different types of network architectures, is referred to as an internetwork. intranet—An intranet is an internal network. An intranet is generally only accessible by corporate network users, but it runs all the same protocols and services as does an Internet site.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

5

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP loopback—The TCP/IP loopback function allows a network administrator to test IP software without concern for the hardware or drivers. The loopback address 127.0.0.1 is the designated software loopback interface for the machine. Medium Access Control (MAC)—MAC is one of the media-specific IEEE 802 standards (802.3, 802.4, and 802.5) that defines the protocol and frame formats for Ethernet, Token Bus, and Token Ring. It is the lower sublayer of the Data Link Layer of the OSI model used to transmit frames between NICs. network interface card (NIC)—A NIC is an expansion board inserted into a computer to enable the computer to be connected to a network. Network News Transfer Protocol (NNTP)—NNTP is the Internet protocol used to transfer news messages between news servers and clients. Users use news clients to post, distribute, and retrieve news articles from a database stored on a central news server. Open Systems Interconnection (OSI)—OSI began as a reference model, that is, an abstract model for data communications. However, now the OSI model has been implemented and is used in some data communications applications. The seven-layer OSI model falls logically into two parts. Layers 1 through 4, the “lower” layers, are concerned with the communication of raw data. Layers 5 through 7, the “higher” layers, are concerned with the networking of applications. PC card—Previously known as a PCMCIA card, a PC card is a small, rectangular expansion card for portable devices, such as laptop and palm computers. PC cards can be NICs, modems, and hard drives, to name a few. Post Office Protocol (POP)—POP is used to transfer information from a mail server to a user’s computer so the information can be read by a mail program at the user’s desk. POP3 is the latest iteration of the protocol. protocol data unit (PDU)—The concept of a PDU is used in the OSI reference model. From the perspective of a protocol layer, a PDU consists of information from the layer above plus the protocol information appended to the data by that layer. For example, a frame is a PDU of the Data Link Layer, and a packet is a PDU of the Network Layer.

6

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Overview repeater—A repeater connects one cable segment of a LAN to other segments, including connecting differing media. For example, a repeater connects thin Ethernet cables to thick Ethernet cables. It regenerates electrical signals from one segment of a cable onto all other segments. Because a repeater reproduces exactly what it receives, bit by bit, it also reproduces errors. Request for Comment (RFC)—RFCs are the working documents of the Internet research and development community. A document in this series may be on essentially any topic related to computer communication and may be anything from a meeting report to the specification of a standard. router—A router is a Layer 3 device with several ports that can each connect to a network or another router. The router examines the logical network address of each packet, then uses its internal routing table to forward the packet to the routing port associated with the best path to the packet’s destination. If the packet is addressed to a network that is not connected to the router, the router forwards the packet to another router that is closer to the final destination. Each router, in turn, evaluates each packet and then either delivers the packet or forwards it to another router. Simple Network Management Protocol (SNMP)—SNMP is a TCP/ IP Application Layer protocol used to send and receive information about the status of network resources on a TCP/IP network. switch—A switch is a device that operates at the Data Link Layer of the OSI reference model. It can connect LANs or segments of the same media access type and dedicate its entire bandwidth to each frame it switches. Systems Network Architecture (SNA)—SNA is IBM’s architecture for computer networking, which was designed for transaction processing in mission-critical applications. SNA networks usually involve a large number of terminals communicating with a mainframe. Telnet—Telnet is a TCP/IP Application Layer protocol that provides a remote login capability to another computer on a network. Token Ring—Token Ring is the IEEE 802.5-specified, ring-based, token-passing LAN topology. Each node on the ring acts as a repeater, passing a token from node to node as the token travels around the entire ring. Each node must wait its turn to transmit data, and may only transmit when it controls the token.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

7

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP Transmission Control Protocol (TCP)—TCP is the TCP/IP Transport Layer protocol that provides reliable, full-duplex, stream service. TCP allows a process on one computer to send data to a process on another computer. TCP software implementations normally reside in the OS and use IP to transmit information across the underlying internet. Trivial File Transfer Protocol (TFTP)—TFTP is the TCP/IP protocol for file transfer with minimal capability and overhead. TFTP depends on the unreliable, connectionless, datagram delivery service UDP. TFTP is designed for use on diskless workstations that keep such software in ROM. unicast—A unicast is a transmission sent to a single network address. This is in contrast to a broadcast, which is sent to all network addresses simultaneously, and a multicast, which is sent to several addresses at once. Universal Serial Bus (USB) adapter—A USB adapter is a PC expansion device that conforms to the USB standard. USB devices support data transfer rates of up to 12 Mbps and up to 127 devices on the same bus. USB device examples include NICs, modems, scanners, and CD-ROM drives. UNIX-to-UNIX Copy Program (UUCP)—UUCP is a standard UNIX utility that copies files between UNIX systems. It can be used for e-mail transfer. User Datagram Protocol (UDP)—UDP is the TCP/IP protocol that allows an application program on one computer to send a datagram to an application program on another computer. UDP uses IP to deliver datagrams. The difference between UDP datagrams and IP datagrams is that UDP includes a protocol port number, allowing the sender to distinguish among multiple destinations (application programs) on the remote computer. UDP also includes a checksum for the data being sent.

8

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Lesson 1—OSI Model Review

Lesson 1—OSI Model Review This lesson reviews some of the OSI model’s key principles. The OSI model is used as a frame of reference in almost all computer networking texts and courses, and in this course, as well. The information reviewed in this lesson is critical to understanding computer networking.

Objectives At the end of this lesson you will be able to: •

Describe each OSI model layer’s primary function

•

List the two main address types found in most computer networks

•

Explain why three address types are used to move information from source to destination

Key Point

There are two main types of computer addresses: physical and logical.

Primary Functions of the OSI Model Layers The OSI Model Layers Table provides an overview of each OSI model layer’s primary functions. It also presents the layer’s unit of information and address type where appropriate. OSI Model Layers OSI Model Layer

Layer Function

Unit of Information

Application

User functionality

Program

Presentation

Character representation Compression Security

Characters and words

Session

Establishing, conducting, and ending sessions

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Address Type

9

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP OSI Model Layers (Continued) OSI Model Layer

Physical Addresses

Layer Function

Unit of Information

Address Type

Transport

Transmitting messages from sending computer process to receiving computer process

Message

Process to process between applications

Network

Transmitting individual packets across a network

Packet

Packet address identifying receiver’s network and host location

Data Link

Transmitting frames containing a packet across a link en route to final destination

Frame

NIC (next node in network)

Physical

Transmitting bits in the form of signals across physical media

Bit

A physical address is also referred to as a: •

Hardware address

•

Adapter address

•

Network interface card (NIC) address

•

Medium Access Control (MAC) address

A physical address is required for network devices to ultimately deliver information to a given network node. The word “ultimately” is used because information often starts out (at the higher layers) addressed to some symbolic name, such as the host name in the command “ftp serverhost.” The name “serverhost” refers to the name of the target host computer the user is attempting to contact using the ftp (TCP/IP) application and protocol. For the user to connect to this host, a physical address must somehow be derived from the symbolic name, and then used in an agreedupon addressing scheme to reach the target. In this case, a name service such as the Domain Name Service (DNS) derives a logical software address from the symbolic name. The TCP/IP protocol suite calls this logical address an IP address.

10

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Lesson 1—OSI Model Review

The Physical Layer

You may want to associate a physical address with the OSI model Physical Layer; indeed, this seems natural. However, the OSI model Data Link Layer actually processes the physical address. The Physical Layer only transmits bits to, and receives bits from, the physical medium, and does not “see” the bits as organized into meaningful patterns, such as an address. The Physical Layer operates depending on the chosen network topology. For example, Ethernet networks format, frame, and time bits in a different manner than do Token Ring networks.

The Data Link Layer

We can categorize physical addresses, for the purposes of networking, into two general types: local area network (LAN) addresses and wide area network (WAN) addresses. A LAN address is commonly found in an Ethernet or Token Ring LAN environment, while we use WAN addresses in High-Level Data Link Control (HDLC) or frame relay network protocol addressing. Physical addresses support a flat network model. This means that physical addresses are not routable; they only designate a host, not both a network and host. To be routable, as in TCP/IP, a protocol must designate both a host and network address. Devices communicating with a Layer 2 only addressing scheme cannot communicate across network (Layer 3) boundaries.

Logical Addresses

A logical address differs from a physical address in that it is generally implemented as a software entity rather than a hardware entity. There are two primary types of logical addresses, as follows:

The Network Layer

•

Network addresses, processed at the Network Layer

•

Port or process addresses, processed at the Transport Layer

One example of a logical address is an IP address (network), such as 144.25.54.8. The network administrator assigns each host on an IP network a unique IP address. The IP address is a 32-bit address that designates both a host portion and a network portion.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

11

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP Because layer protocols designate both a host and a network, they allow us to build hierarchical networks. The Three-Layer Network Hierarchy Diagram illustrates this concept.

Router

WAN

Router

Core

Router Router Router

Distribution Router

Hub Switch

Access

Hub Switch Hub Switch

Three-Layer Network Hierarchy Layer 3 addresses allow us to create multiple routed networks. We can use the three-layer hierarchical model to isolate traffic to local network segments, only passing traffic between layers when hosts on one segment must communicate with hosts on other segments. Ideally, we would isolate traffic to the bottom two layers, the access and distribution layer, and only pass WAN traffic to the core layer. Routing and Layer 3 addresses make this possible.

12

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Lesson 1—OSI Model Review

The Transport Layer

Another logical address example is a port number (process address), such as “23.” Devices use port addresses to pass information to the higher layers. Devices use port numbers to keep track of multiple simultaneous conversations. Network applications use well-known port numbers to communicate with their peer processes. For example, Web browsers request Web services on the well-known TCP Port 80. The Well-Known Port Numbers Table lists some of the more commonly used TCP and User Datagram Protocol (UDP) addresses. Well-Known Port Numbers Decimal

Protocol-Description

20

TCP-FTP Data

21

TCP-FTP

23

TCP-Telnet

25

TCP-SMTP

53

TCP/UDP-DNS

67

UDP-BOOTP/DHCP

69

UDP-TFTP

161

UDP-SNMP

The Transport Layer is responsible not only for application addressing, but also for providing reliable communications over the best effort Layer 3 protocols. The Transport Layer provides flow control, windowing, data sequencing, and recovery.

Layers 5 Through 7

The remaining OSI model layers work with the data itself. These layers leave the end-to-end data transport issues to the lower four layers.

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

13

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP

The Session Layer

The Session Layer is responsible for establishing, conducting, and terminating conversations (sessions) between application entities. The Session Layer provides the following services to the higher layer protocols: •

Establish a session

•

Conduct dialogs to prevent each conversing node from transmitting data simultaneously

•

Manage session activities by dividing each conversation into parts

•

End the session gracefully

The Presentation Layer

The Presentation Layer is responsible for data representation, security, and compression. This layer ensures that the sender transfers information to the receiver in a format that the receiving application can understand. The Presentation Layer adds data encryption and decryption protocols to protect data from unauthorized access and, by relying on data compression protocols, ensures efficient use of the communications channel.

The Application Layer

The Application Layer provides the user interface to the system. Not all applications are user applications, however, with some designed to provide services to other applications. A network application identifies its peers, establishes communications, and enacts procedures for peer-to-peer data transfer. The peer applications may also determine whether sufficient network resources exist for successful communications.

Addressing Summary The most important fact to remember concerning logical addresses is that a logical address will not transmit information “into the box.” Only the physical address, whether it is a broadcast, multicast (group), or single destination (unicast) address, can accomplish this.

14

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

Copyrighted Material for Preview Use Only Lesson 1—OSI Model Review

Activities 1. The OSI model Physical Layer serves which primary purpose? a.

It provides physical addresses for moving information into the box.

b. It builds frames that carry higher layer PDU across network segments. c.

It uses network and host addresses to route packets across networks.

d. It transmits data in the form of bits across the physical medium. 2. Which is the primary purpose of the OSI model Session Layer? a.

It builds frames that carry higher layer PDU across network segments.

b. It builds a flat network model with unroutable physical addresses. c.

It establishes, manages, and terminates dialogs between Presentation Layer entities.

d. It defines the format in which Application Layer information is sent across the network. 3. Which is the primary purpose of the OSI model Presentation Layer? a.

It establishes, manages, and terminates sessions between applications.

b. It uses logical addresses to identify higher layer applications. c.

It builds a flat network model with unroutable physical addresses.

d. It defines the format in which Application Layer information is sent across the network. 4. The OSI model Data Link Layer serves which primary purpose? a.

It builds a flat network model with unroutable physical addresses.

b. It transmits data in the form of bits across the physical medium. c.

It creates multiple routed networks with logical addresses.

d. It establishes, manages, and terminates sessions between applications. ©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902

15

Copyrighted Material for Preview Use Only Unit 1 — Introduction to TCP/IP 5. The OSI model Application Layer serves which primary purpose? a.

It defines well-known logical port addresses for communications between applications.

b. It controls dialogs between two or more Presentation Layer entities. c.

It establishes user application procedures for error recovery and data integrity.

d. It creates a hierarchical addressing model for moving information across networks.

Extended Activities 1. On a Windows PC, select Start, Find, Files, or Folders. In the named box, type services. Select the drive on which the operating systems is installed (typically C:) in the Look In box. In the Results window, double-click the Services file (there is no extension). When prompted for a program with which to open the file, choose Notepad. Notepad will open the file; this file contains all the TCP and UDP well-known service ports as defined by RFC 1060. 2. Locate in the Services file the following services and list their port numbers: a.

BOOTP

b. Chargen c.

POP3

d. NNTP e.

16

Nbname (NetBIOS name)

©2003 WestNet Learning www.westnetlearning.com 1-888-452-6902