1

NETWORKING BASICS

Chapter 1: NETWORKING BASICS

LANS, WANS, AND MANS

Three main networking technologies are used to connect computers and networks together:

Local area network (LAN) Wide area network (WAN)

Metropolitan area network (MAN)

Chapter 1: NETWORKING BASICS

LAN

Used to network computers located in a limited geographical area such as a room, floor, or building Has three main attributes

Topology: bus, star, and ring

Medium: copper, fiber, and wireless Protocols

Chapter 1: NETWORKING BASICS

EXAMPLE OF A LAN

Chapter 1: NETWORKING BASICS

WAN

Spans a large geographic area Defined as an internetwork (private or public) that connects many LANs Uses routers and usually point-to-point links Examples of WANs

The Internet Private global enterprise networks

NETWORKING BASICS

EXAMPLE OF A WAN

Storage-Area Networks (SANS)

Virtual Private Networks (VPNs)

Benefits of VPNs

Intranet and Extranet VPN

Importance of Bandwidth

Bandwidth Pipe Analogy

Bandwidth Highway Analogy

Chapter 1: NETWORKING BASICS

14

PROTOCOLS AND SERVICES

Facilitate communication between two networked devices Perform data encapsulation Provide a common language to communicating devices Provide either connectionless or connectionoriented services

CONNECTIONLESS VS. CONNECTIONORIENTED

Chapter 1: NETWORKING BASICS

15

Connectionless protocols

Do not require a connection Have very little overhead Are fast and unreliable Require a connection

Connection-oriented protocols

Require more overhead

Are slower and very reliable

Chapter 1: NETWORKING BASICS

16

PHYSICAL LAYER TRANSMISSION

Chapter 1: NETWORKING BASICS

17

NETWORK LAYER ADDRESSING

Chapter 1: NETWORKING BASICS

18

ROUTING

Chapter 1: NETWORKING BASICS

19

SUMMARY

LANs, WANs, and MANs allow groups of computers to share information. Signaling protocols are used to transmit data as 1s and 0s.

Protocols provide a common language for communication hosts.

20

Chapter 2

NETWORK CABLING

Chapter 2: NETWORK CABLING

21

TOPOLOGIES

There are three main local area network (LAN) topologies:

Bus Star

Ring
Mesh Wireless

Other network topologies include:

Chapter 2: NETWORK CABLING

22

BUS TOPOLOGY

The bus topology supports thick and thin coaxial segments.

Segments are connected by repeaters.

The bus topology uses the baseband signaling method. Signals are broadcast in both directions simultaneously.

Both ends of each segment require termination to avoid reflection.

End systems connect to the segment in a linear manner.

Chapter 2: NETWORK CABLING

23

THICK AND THIN COAXIAL BUS

Chapter 2: NETWORK CABLING

24

STAR TOPOLOGY

The star topology can use coaxial, twisted pair, or fiber optic cable. A central device (hub) connects hubs and nodes to the network.

Each node connects to its own dedicated port on the hub. Hubs broadcast transmitted signals to all connected devices. You can connect multiple hubs to form a hierarchical star topology.

The star topology uses the baseband signaling method.

Chapter 2: NETWORK CABLING

25

A SIMPLE STAR TOPOLOGY

Chapter 2: NETWORK CABLING

26

A HIERARCHICAL STAR TOPOLOGY

Chapter 2: NETWORK CABLING

27

RING TOPOLOGY

The ring topology can use twisted pair or fiber optic cabling.

A central device (hub) connects hubs and nodes to the network.

Each node connects to its own dedicated port on the hub.

You can connect multiple hubs to form a larger ring.

The ring topology uses the baseband signaling method. Frames are transmitted around the ring from node to hub to node. Media Access Control (MAC) is used for token passing.

Chapter 2: NETWORK CABLING

28

A RING NETWORK

Chapter 2: NETWORK CABLING

29

MESH TOPOLOGY

Typically uses fiber optic cabling for redundant wide area network (WAN) links Provides multiple paths to destinations for fault tolerance

Supports baseband and broadband signals

Requires an enormous amount of cable

Chapter 2: NETWORK CABLING

30

LAN MESH

Chapter 2: NETWORK CABLING

31

ENTERPRISE MESH

Chapter 2: NETWORK CABLING

32

WIRELESS TOPOLOGY

Cell-based technology that uses unbounded media Two wireless topologies:

Ad hoc Infrastructure

Chapter 2: NETWORK CABLING

33

AD HOC WLAN

Chapter 2: NETWORK CABLING

34

INFRASTRUCTURE WLAN

Chapter 2: NETWORK CABLING

35

LAN CABLE TYPES

Three cable types are used in LANs:

Coaxial Twisted pair Fiber optic

Chapter 2: NETWORK CABLING

36

COAXIAL CABLE

Chapter 2: NETWORK CABLING

37

AUI (ATTACHMENT UNIT INTERFACE) CABLE

Chapter 2: NETWORK CABLING

38

THIN ETHERNET HARDWARE

Chapter 2: NETWORK CABLING

39

UNSHIELDED TWISTED PAIR (UTP) CABLE

Chapter 2: NETWORK CABLING

40

UTP CONNECTORS

Chapter 2: NETWORK CABLING

41

UTP CABLE GRADES

Category
1 2

Frequency
Up to 0 MHz Up to 1 MHz

Primary Application
Voice networks Voice and low-speed data networks less than 4 Mbps

3
4 5

Up to 16 MHz
Up to 20 MHz Up to 100 MHz

Voice and data networks from 4 to 100 Mbps

16-Mbps Token Ring 100-Mbps Fast Ethernet

5e 6

Up to 100 MHz Up to 250 MHz

1000-Mbps Gigabit Ethernet 1000-Mbps Gigabit Ethernet

Chapter 2: NETWORK CABLING

42

FIBER OPTIC CABLE

Chapter 2: NETWORK CABLING

43

STRAIGHT TIP (ST) CONNECTOR

Chapter 2: NETWORK CABLING

44

STRAIGHT THROUGH AND CROSSOVER WIRING

Wiring within a twisted pair cable is configured as

either

Straight through, where each wire (or pin) is attached

to the same contact point at each end

Crossover, where transmit contacts on each end of

the cable are connected to the receive contact at the other end

Chapter 2: NETWORK CABLING

45

STRAIGHT THROUGH WIRING

Chapter 2: NETWORK CABLING

46

CROSSOVER WIRING

Chapter 2: NETWORK CABLING

47

SUMMARY

The three basic LAN topologies are bus, star, and ring. WLANs are becoming more popular. Mesh networks are not typically used in LANs. The primary cable types used in LANs are coaxial, twisted pair, and fiber optic.

48

Network Devices

Network Interface Card

NIC
Ethernet network interface card (NIC), which inserts into the system bus of a computer and makes the connection between running software processes on the computer and physical media.

Repeater

A repeater

is a device that amplifies a signal, to

counteract the effects of attenuation

HUB

HUB
is a device used to connect all of the computers on a
star network.
From the outside, a hub looks like nothing more than

a box with a series of cable connectors and LEDs in it

Two Types: Active and Passive

Active :amplifies the signals and immediately

transmits them through all of the other ports. ports .

Passive: transmits the signals through all of the other

Communicating Within the LAN

Communicating Within the LAN

Broadcast: message is a packet with a special

destination address that causes it to be read and processed by every computer that receives it computer on the network

Unicast: message is a packet addressed to a single Multicast: message is addressed to a subset of

computers on the network

LAN Segment Limitations

Signals degrade with transmission distance. Each Ethernet type has a maximum segment length.

Extending LAN Segments

Bandwidth is shared
Extends cable distances Repeats or amplifies signal

Collisions

Bridges

Operate at Layer 2 of the OSI model Forward, filter, or flood frames Few ports Slow

Layer 2 devices
A layer 2 device is a device that understand MAC, for example: NIC (Network Interface Card) Bridge : - address learning - forwarding decisions are based on software - bridge is used for LAN segmentation - max. 16 port. Switch: - a multi-port bridge up to 567 port - forwarding decisions are based on hardware ASIC (faster than bridge)

LANs Today

Users grouped by physical location

Many switches Switches connected by high-speed links

Multiple Collision Domains

MAC Address Components

MAC Addresses

Decimal, Binary, and Hexadecimal Number Systems

Binary and Hexadecimal Number Comparisons

Decimal-to-Binary-Conversion

Binary-to-Hexadecimal Conversion Example

Hexadecimal-to-Binary Conversion Example

Examples

Convert the following:

205
10

to binary
to binary

1100110 1 0000101
0 226 199 8C E0 134

11100010 to decimal
11000111 to decimal

10001100 to hexadecimal
224 to hexadecimal

0x86

to decimal

Router

functions of a router
- connect between networks - Select best path - Divide broadcast domain

- Packet forwarding

collision domain & broadcast domain

broadcast domain :is the group of computers that

receive a broadcast message transmitted by any one of the computers in the group.

A collision domain is a network (or part of a

network) that is constructed so that when two computers transmit packets at the same time, a collision occurs, causing both packets to be lost. All hosts that are affected by a collision belong to the same collision domain.

Device hub Switch router

Collision domain
one Equal number of ports Equal number of ports

Broadcast Domain
one one Equal number of ports

How many collision domains and broadcast domains ?

collision domains = 4 & Broadcast domains =1

Identifying Collision and Broadcast Domains

Collision domain =3 and Broadcast

UTP Implementation (Straight-Through)

Cable 10BASE-T/ 100BASE-TX Straight-Through Straight-Through Cable

Pin Label 1 2 3 4 5 6 7 8 TX+ TXRX+ NC NC RXNC NC 1 2 3 4 5 6 7 8

Pin Label TX+ TXRX+ NC NC RXNC NC

Wires on cable ends are in same order.

UTP Implementation (Crossover)

Cable 10BASE-T or 100BASE-TX Straight-Through Crossover Cable

Pin Label 1 2 3 4 5 6 7 8 TX+ TXRX+ NC NC RXNC NC 1 2 3 4 5 6 7 8

Pin Label TX+ TXRX+ NC NC RXNC NC

EIA/TIA T568A

EIA/TIA T568B

Some wires on cable ends are crossed.

UTP Implementation: StraightThrough vs. Crossover

Using Varieties of UTP

Crossover

Crossover

Straight-through

Straight-through

Straight-through

Multi-Layer Switch

High port density Large frame buffers

Mixture of port speeds

Fast internal switching

83

Chapter 4

DATA-LINK LAYER PROTOCOLS

Ethernet Token Ring FDDI Wireless Networking

Chapter 4: DATA-LINK LAYER PROTOCOLS

84

OSI MODEL AND IEEE 802.X STANDARDS

The Institute of Electrical and Electronics Engineers (IEEE)

divides the data-link and physical layers of the Open Systems Interconnection (OSI) model into four sublayers:
Two IEEE data-link sublayers:
The upper sublayer is the 802.2 or Logical Link Control (LLC)
Operates independently from the Media Access Control (MAC)

sublayer

Specifies the upper layer protocol carried within a frame

The lower sublayer is the MAC sublayer.

Defined by various 802.x standards such as IEEE 802.3, 802.5, and

802.11

OSI MODEL AND IEEE 802.X STANDARDS (CONT.)

Two IEEE physical sublayers:
Physical signaling
Media specifications

Chapter 4: DATA-LINK LAYER PROTOCOLS

85

Chapter 4: DATA-LINK LAYER PROTOCOLS

86

FOUR IEEE SUBLAYERS

Chapter 4: DATA-LINK LAYER PROTOCOLS

87

IEEE DATA-LINK AND PHYSICAL STANDARDS

Chapter 4: DATA-LINK LAYER PROTOCOLS

88

ETHERNET VERSION I AND II

Digital Equipment Corporation, Intel, and Xerox
Version I (also known as DIX Ethernet)
Uses bus topology with RG-8 (thick coaxial cable)

published the first two 10-Mbps Ethernet standards.

Uses Attachment Unit Interface (AUI) patch cables with

15-pin connectors, 50-ohm terminators, and external transceivers Maximum distance per segment: 500 meters
Maximum distance per network: 2500 meters
Uses the 5-4-3 rule

Adopted and renamed 10Base5 (page 157) by the IEEE

Chapter 4: DATA-LINK LAYER PROTOCOLS

89

ETHERNET VERSION I AND II (CONT.)

Version II
Uses bus topology with RG-58 (thin coaxial cable)
Uses T-connectors, 50-ohm terminators, and internal

transceivers

Maximum distance per segment: 185 meters

Maximum distance per network: 925 meters Uses the 5-4-3 rule

Adopted and renamed 10Base2 by the IEEE

Chapter 4: DATA-LINK LAYER PROTOCOLS

90

5-4-3 RULE

Chapter 4: DATA-LINK LAYER PROTOCOLS

91

IEEE ETHERNET STANDARDS (page 159)

The IEEE adopted and renamed the original

Ethernet standards and then expanded them.

802.3 working group.

All IEEE Ethernet standards are controlled by the

10-Mbps standards: 802.3a (10Base2), 802.3e

(10Base5), 802.3i (10Base-T), and 802.3j (10BaseFP, 10Base-FB, and 10Base-FL)

100-Mbps standards: 802.3u (100Base-X) 1000-Mbps standards: 802.3z and 802.3ab

(1000Base-X)

Chapter 4: DATA-LINK LAYER PROTOCOLS

92

IEEE 10BASE-X STANDARDS

The three primary IEEE standards for 10-Mbps

baseband networks are

10Base5

Physical and data-link layer standards and limitations

are identical to Ethernet version I.

10Base2
Physical and data-link layer standards and limitations

are identical to Ethernet version II.

Chapter 4: DATA-LINK LAYER PROTOCOLS

93

IEEE 10BASE-X STANDARDS (CONT.)

10Base-T
Star topology using unshielded twisted-pair (UTP)

cabling Two-pair UTP with RJ-45 connectors: One pair for transmit, the other one for receive
Supports half-duplex and full-duplex modes Maximum distance per segment: 100 meters

Maximum distance per network: 500 meters (which

includes connections from workstation to hub and also connections between hubs) Uses the 5-4-3 rule

Chapter 4: DATA-LINK LAYER PROTOCOLS

94

10BASE-T 5-4 Rule

Chapter 4: DATA-LINK LAYER PROTOCOLS

95

IEEE 100BASE-X STANDARDS

The three IEEE standards for 100-Mbps baseband

networks configured as a star topology are

100Base-TX

Uses the 4B5B encoding scheme over two pair

(Category 5), the same as 10Base-T

Supports half-duplex mode or full-duplex mode Maximum distance per segment (half or full): 100

meters

Maximum distance per half-duplex network: 205

meters Supports Class I and Class II repeaters

Chapter 4: DATA-LINK LAYER PROTOCOLS

96

IEEE 100BASE-X STANDARDS (CONT.)

100Base-T4
Uses the 8B/6B encoding scheme over four Category 3

(CAT3) twisted pairs

Supports only half-duplex mode

Maximum distance per segment: 100 meters

Maximum distance per network: 205 meters Supports Class I and Class II repeaters

Chapter 4: DATA-LINK LAYER PROTOCOLS

97

IEEE 100BASE-X STANDARDS (CONT.)

100Base-FX
Uses the 4B/5B encoding scheme over fiber optic
Supports half-duplex mode or full-duplex mode
Maximum length of a multimode half-duplex segment: 412

meters Maximum length of a multimode full-duplex segment: 2 kilometers Maximum length of a singlemode half-duplex segment: 2 kilometers Maximum length of a singlemode full-duplex segment: 10+ kilometers

Supports Class I and Class II repeaters

Chapter 4: DATA-LINK LAYER PROTOCOLS

98

CLASS I AND CLASS II REPEATERS

Chapter 4: DATA-LINK LAYER PROTOCOLS

99

IEEE 1000BASE-X STANDARDS

The IEEE 1000Base-X standard defines Gigabit

Ethernet specifications for twisted-pair cable and fiber optic cable.

Uses the 8B/10T encoding scheme

Supports full-duplex mode only

Maximum length of UTP segment: 100 meters Maximum length of multimode fiber segment: 220+

meters

Maximum length of singlemode fiber segment: 5000

meters

Chapter 4: DATA-LINK LAYER PROTOCOLS

100

FOUR ETHERNET FRAME TYPES

There are four different Ethernet frame types:
Version II Ethernet 802.3 IEEE 802.3 IEEE 802.3 SNAP

Chapter 4: DATA-LINK LAYER PROTOCOLS

101

VERSION II FRAME

Chapter 4: DATA-LINK LAYER PROTOCOLS

102

ETHERNET 802.3 FRAME

Chapter 4: DATA-LINK LAYER PROTOCOLS

103

IEEE 802.3 FRAME

Chapter 4: DATA-LINK LAYER PROTOCOLS

104

IEEE 802.3 SNAP FRAME

Chapter 4: DATA-LINK LAYER PROTOCOLS

105

MAC ADDRESSES (page 162)

Media Access Control: Is the mechanism that enables multiple computers to use the same network medium without conflicting

Chapter 4: DATA-LINK LAYER PROTOCOLS

106

CARRIER SENSE MULTIPLE ACCESS WITH COLLISION DETECTION (CSMA/CD)

Phase Carrier sense Description A computer listens to the network before transmitting. When the network is clear, the computer transmits the packet. The computer checks for signs of a collision. If one occurs, it retransmits the packet.

Multiple access

Collision detection

CARRIER SENSE MULTIPLE ACCESS WITH COLLISION DETECTION (CSMA/CD)

All half-duplex implementations of Ethernet use the

Chapter 4: DATA-LINK LAYER PROTOCOLS

107

CSMA/CD channel access method.

Carrier Sense

A device that wants to transmit must first listen to the

channel to see if it is in use.

If the channel is busy, the device must wait. If the channel is idle, the device can transmit a frame.

CARRIER SENSE MULTIPLE ACCESS WITH COLLISION DETECTION (CSMA/CD) (CONT.)

Multiple Access

Chapter 4: DATA-LINK LAYER PROTOCOLS

108

CSMA All devices on the network contend for access to the channel.
When two or more devices transmit at the same time,

Collision Detection

their signals collide.

Devices detect collisions when they receive a different

frequency on their receive pair.

Devices must immediately stop transmitting data and

send out a jamming signal and then back off for a random interval before trying again.

Chapter 4: DATA-LINK LAYER PROTOCOLS

109

COLLISIONS

Collision

Contention

Collisions are also called signal quality errors.

They are normal on Ethernet networks. The frequency of collisions increases as network

traffic increases.

Late collisions are a sign of a serious problem.

Do Exercise 4-2 (Page 194)

Chapter 4: DATA-LINK LAYER PROTOCOLS

110

TOKEN RING

Token Ring Network

Token Ring was originally developed by IBM, and then it was

adopted by the IEEE and renamed 802.5.

Star wired ring topology Operates at either 4 Mbps or 16 Mbps

Token-passing channel access method (Next Slide) Uses Multistation Access Units (MAUs) to connect nodes to the

network

You can connect MAUs together, using RI (Ring In) and RO (Out)

ports to form a larger ring.

Can use both shielded and unshielded twisted-pair cable

Chapter 4: DATA-LINK LAYER PROTOCOLS

111

TOKEN Passing

Token Passing

A token frame circulates continuously around the

network.
data.

Only the computer holding the token can transmit

The transmitting system is responsible for removing

the data from the ring.

Chapter 4: DATA-LINK LAYER PROTOCOLS

112

TOKEN RING FRAMES

Chapter 4: DATA-LINK LAYER PROTOCOLS

113

FDDI (Fiber Distributed Data Interface )

Developed by the American National Standards Institute

(ANSI)

Uses dual ring topology

The primary ring serves as a data path. The secondary ring provides fault tolerance.

Has a 100-Mbps transmission rate over fiber optic cabling Uses the token passing channel access method Supports early token release Uses single attachment station (SAS) or dual attachment

station (DAS) FDDI

Chapter 4: DATA-LINK LAYER PROTOCOLS

114

FDDI (Fiber Distributed Data Interface )

Supports both:
Singlemode cable (600 Km segments) Multimode cable (100 Km segments &

500 Workstations) This is the industry standard for fiber optic LANs.

Chapter 4: DATA-LINK LAYER PROTOCOLS

115

SAS AND DAS DEVICES

Chapter 4: DATA-LINK LAYER PROTOCOLS

117

RING WRAP (page 185)

Chapter 4: DATA-LINK LAYER PROTOCOLS

118

FDDI FRAMES

Chapter 4: DATA-LINK LAYER PROTOCOLS

119

WIRELESS LANS

Wireless LANs

IEEE standard 802.11 defines the specifications for wireless

LANs (WLANs).

Support various transmission rates, depending on the standard

802.11b supports up to 11 Mbps. 802.11a and 802.11g support up to 54 Mbps.

Support ad hoc or infrastructure topologies Use three different signaling methods: Direct Sequence Spread

Spectrum (DSSS), Frequency Hopping Spread Spectrum (FHSS), and infrared (CSMA/CA) channel access method

Use the Carrier Sense Multiple Access with Collision Avoidance

Chapter 4: DATA-LINK LAYER PROTOCOLS

120

AD HOC WLAN

Chapter 4: DATA-LINK LAYER PROTOCOLS

121

INFRASTRUCTURE WIRELESS

Chapter 4: DATA-LINK LAYER PROTOCOLS

122

SUMMARY
DIX and IEEE 802.3 define physical and data-link

layer standards and functions for Ethernet networks using CSMA/CD over coaxial, twisted-pair, or fiber optic cabling. layer standards and functions for a token passing ring topology.
standards for a token-passing, fiber optic ring topology.

IBM and IEEE 802.5 define physical and data-link FDDI defines the physical and data-link layer

WLANs can either be ad hoc or infrastructure.

123

Chapter 5

NETWORK LAYER PROTOCOLS

IP IPX NetBEUI AppleTalk

Chapter 5: NETWORK LAYER PROTOCOLS

124

EXAMPLES OF NETWORK LAYER PROTOCOLS

The Transmission Control Protocol/Internet Protocol

(TCP/IP) suite

Internet Protocol (IP)

Novells Internetwork Packet Exchange/Sequenced Packet

Exchange (IPX/SPX) suite

Internetwork Packet Exchange (IPX)

Apple Computers AppleTalk suite

Datagram Delivery Protocol (DDP)

Microsofts suite
NetBIOS Extended User Interface (NetBEUI)

Chapter 5: NETWORK LAYER PROTOCOLS

125

THE INTERNET PROTOCOL

IP, defined in Request for Comments (RFC) 791, is a

connectionless network layer protocol that provides

Datagram encapsulation Logical addressing

Fragmentation and reassembly of datagrams

Routing

Chapter 5: NETWORK LAYER PROTOCOLS

126

IP FUNCTIONS
Encapsulation IP forms a datagram by adding an IP header to information

passed down from the transport layer protocol.

Addressing Each datagram includes logical source and destination

addresses.

Fragmentation and reassembly

The source host or router divides packets into smaller

datagrams that can be transmitted over the network.

The destination host reassembles fragments when it receives

them.

Routing
The selection of the most efficient path.

Chapter 5: NETWORK LAYER PROTOCOLS

127

DATAGRAM ENCAPSULATION

Chapter 5: NETWORK LAYER PROTOCOLS

128

FRAGMENTATION AND REASSEMBLY

Routers connect networks that support different-sized

packets.

The largest packet size supported by a network is called its

maximum transmission unit (MTU).

When a packet is too large to be forwarded to a particular

network, the router splits it into fragments.

Chapter 5: NETWORK LAYER PROTOCOLS

129

FRAGMENTATION AND REASSEMBLY

Each fragment is encapsulated with a header and is

transmitted as a separate packet.

Fragments are not reassembled until they reach their final

destination.

Fragments can themselves be fragmented.

Fragmentation

Chapter 5: NETWORK LAYER PROTOCOLS

130

FRAGMENTATION AND REASSEMBLY

Chapter 5: NETWORK LAYER PROTOCOLS

132

IP HEADER AND FIELDS

Chapter 5: NETWORK LAYER PROTOCOLS

134

Protocol Field Values (The most commonly used values)

0
1 3 6 8 17

IP
ICMP Gateway-to-Gateway Protocol (GGP) TCP (most expected) Exterior Gateway Protocol (EGP) UDP (most expected)

Chapter 5: NETWORK LAYER PROTOCOLS

136

IP ADDRESSING
IP addresses are
Logical network layer addresses used to identify

networks, subnetworks, and hosts

4 bytes (or 32 bits) in length and represented in

dotted decimal notation

The values within each byte range from 0 to 255.

Public or private

Chapter 5: NETWORK LAYER PROTOCOLS

137

DECIMAL AND BINARY NUMBERING

The decimal numbering system uses 10 (base 10)

values to represent numbers.

Uses 09

The binary numbering system uses 2 (base 2)

values to represent numbers.

Uses 0 and 1

Chapter 5: NETWORK LAYER PROTOCOLS

138

EXAMPLE OF 8-BIT CONVERSION

Chapter 5: NETWORK LAYER PROTOCOLS

139

EXAMPLE OF 16-BIT CONVERSION

Chapter 5: NETWORK LAYER PROTOCOLS

140

THREE IP ADDRESS CLASSES

Chapter 5: NETWORK LAYER PROTOCOLS

141

THREE DEFAULT MASKS

Chapter 5: NETWORK LAYER PROTOCOLS

142

IP ADDRESS CLASSES AND PARAMETERS

Chapter 5: NETWORK LAYER PROTOCOLS

143

EXAMPLE OF A CLASS A ADDRESS

Chapter 5: NETWORK LAYER PROTOCOLS

144

EXAMPLE OF A CLASS B ADDRESS

Chapter 5: NETWORK LAYER PROTOCOLS

145

EXAMPLE OF A CLASS C ADDRESS

Chapter 5: NETWORK LAYER PROTOCOLS

146

IP SUBNETTING
Each address class can be divided further to create

subnets.
bits.

Subnet bits are borrowed from the available host

Class A: 24 host bits Class B: 16 host bits Class C: 8 host bits

Bits used to define subnets cannot be used to

identify hosts.

Borrowed bits are added to the mask.

Chapter 5: NETWORK LAYER PROTOCOLS

147

CLASS A, CLASS B, AND CLASS C SUBNETTING

Chapter 5: NETWORK LAYER PROTOCOLS

150

CLASS C SUBNETTING EXAMPLE (CONT.)

Chapter 5: NETWORK LAYER PROTOCOLS

151

PRIVATE IP ADDRESSES

Chapter 5: NETWORK LAYER PROTOCOLS

152

INTERNET PROTOCOL VERSION 6 (IPV6) ADDRESSING

Addresses the depletion of Internet Protocol version

4 (IPv4) addresses
bits

Increases the address space from 32 bits to 128

Chapter 5: NETWORK LAYER PROTOCOLS

153

INTERNET PROTOCOL VERSION 6 (IPV6) ADDRESSING (CONT.)

Uses six variable-length sections:
Format Prefix Registry ID Provider ID Subscriber ID Subnet ID

Interface ID

Chapter 5: NETWORK LAYER PROTOCOLS

154

EXAMPLE OF AN IPV6 ADDRESS

Chapter 5: NETWORK LAYER PROTOCOLS

156

THE IPX PROTOCOL

Novells IPX protocol is a connectionless network

layer protocol that provides

Datagram encapsulation Logical addressing

Fragmentation and reassembly of datagrams

Routing

Chapter 5: NETWORK LAYER PROTOCOLS

159

IPX ADDRESSING
IPX logical network layer addresses consist of three

parts:

Network
The network portion is four bytes long and is assigned

by an administrator or dynamically during installation.

Node
The node portion is the hardware address of the

interface attached to the network.

Socket
The socket is a two-byte value specifying the

application process.

Chapter 5: NETWORK LAYER PROTOCOLS

160

IPX ADDRESSING

Chapter 5: NETWORK LAYER PROTOCOLS

161

IPX ROUTING PROTOCOLS

There are two routing protocols in the Novell

IPX/SPX suite:

IPX Routing Information Protocol (RIP)

IPX RIP uses broadcasts to learn and advertise routes.

The entire route table is broadcast every 60 seconds.

The maximum number of network hops = 15 (16 is

destination unreachable).

IPX RIP uses two metrics for best path selection: hops

and ticks. (A tick is one-eighteenth of a second.)

Chapter 5: NETWORK LAYER PROTOCOLS

162

IPX ROUTING PROTOCOLS (CONT.)

NetWare Link Services Protocol (NLSP)
NLSP is a link state routing protocol. NLSP does not broadcast; it sends route information

only when there is a change in the network.

The maximum number of hops is 127.

The metric for the best path selection is based on link

parameters, not hops.

THE APPLETALK DATAGRAM DELIVERY PROTOCOL

The DDP protocol is a connectionless network layer

Chapter 5: NETWORK LAYER PROTOCOLS

164

protocol in the AppleTalk suite that provides

Datagram encapsulation Logical addressing

Fragmentation and reassembly of datagrams

Routing

Chapter 5: NETWORK LAYER PROTOCOLS

168

NETBEUI
The NetBEUI protocol was developed by IBM and

then adopted by Microsoft.

delivering NetBIOS data.

NetBEUI is a nonroutable protocol used for NetBEUI does not contain network layer addressing. The NetBEUI frame format includes two

components:

A data-link layer Logical Link Control (LLC) (802.2)

Type II header with control fields A transport layer NetBIOS programming interface

Chapter 5: NETWORK LAYER PROTOCOLS

171

SUMMARY
Network layer protocols like IP, IPX, and DDP

provide data encapsulation, logical addressing, fragmentation, and reassembly.

Class B, and Class C.

There are three classes of IP addresses: Class A, Subnet masks are used to further subdivide Class A,

B, and C networks into subnets.

The NetBEUI protocol is the only network layer

protocol that does not provide logical network layer addressing and is therefore not routable.

172

Chapter 6

TRANSPORT LAYER PROTOCOLS

TCP and UDP SPX and NCP

Chapter 6: TRANSPORT LAYER PROTOCOLS

173

TYPES OF TRANSPORT LAYER PROTOCOLS

There are two types of transport layer protocols:
Connection-oriented
Transmission Control Protocol (TCP) Sequenced Packet Exchange (SPX) NetWare Core Protocol (NCP)

Connectionless
User Datagram Protocol (UDP)

Chapter 6: TRANSPORT LAYER PROTOCOLS

174

TRANSPORT LAYER PROTOCOLS

There are two transport layer protocols in the

Transmission Control Protocol/Internet Protocol (TCP/IP) suite:

TCP

UDP

There are two transport layer protocols in the Novell

Internetwork Packet Exchange (IPX)/SPX suite:

SPX
NCP

Chapter 6: TRANSPORT LAYER PROTOCOLS

175

TRANSMISSION CONTROL PROTOCOL (TCP)

The TCP protocol (described in Request for Comments [RFC]

793) has the following characteristics:

Uses Internet Protocol (IP) ID 06

Is a reliable, connection-oriented protocol Provides guaranteed delivery of packets through sequencing

and acknowledgments

Provides sliding-window flow control Performs error detection and correction

Uses ports to identify the communicating process or application

Chapter 6: TRANSPORT LAYER PROTOCOLS

176

TCP HEADER AND FIELDS (PAGE 253)

Chapter 6: TRANSPORT LAYER PROTOCOLS

178

EXAMPLE OF A TCP HEADER

ESTABLISHING A TCP CONNECTION THREE-WAY HANDSHAKE

Verify that both computers are operating and

Chapter 6: TRANSPORT LAYER PROTOCOLS

179

ready to receive data

Exchange initial sequence numbers (ISNs) Exchange maximum segment sizes (MSSs) Exchange port numbers

Chapter 6: TRANSPORT LAYER PROTOCOLS

180

ESTABLISHING A TCP CONNECTION THREE-WAY HANDSHAKE

TCP Connection

Chapter 6: TRANSPORT LAYER PROTOCOLS

182

TCP CONNECTION TERMINATION

Chapter 6: TRANSPORT LAYER PROTOCOLS

187

USER DATAGRAM PROTOCOL (UDP)

UDP (described in RFC 768) has the following

characteristics:
Uses IP ID 17

Provides fast, connectionless delivery of data

Has less overhead than connection-oriented protocols

Uses ports to identify the communicating process or

application

Chapter 6: TRANSPORT LAYER PROTOCOLS

188

UDP HEADER AND FIELDS (PAGE 266)

Chapter 6: TRANSPORT LAYER PROTOCOLS

190

PORTS & SOCKETS

A port number refers to a specific application or

process running on a computer.

A socket is a combination of a port number and an

IP address. (ex: 192.168.2.10:21) this socket addresses port 21 on the system with address 192.168.2.10

The Internet Assigned Numbers Authority (IANA)

assigns well-known port numbers to common Internet applications.

Chapter 6: TRANSPORT LAYER PROTOCOLS

191

CLIENT AND WELL-KNOWN PORTS

There are two types of TCP and UDP ports:
Client ports

Variable ports with a value from 1024 through

65,534

Server (well-known) ports

Commonly used by applications and services

Port values with a value from 1 through 1023

Chapter 6: TRANSPORT LAYER PROTOCOLS

192

EXAMPLES OF WELL-KNOWN PORTS

TCP ports
File Transfer Protocol (FTP) 20 and 21
Telnet 23 Simple Mail Transfer Protocol (SMTP) 25

UDP ports
Trivial File Transfer Protocol (TFTP) 69 Domain Name System (DNS) 53 Bootstrap Protocol/Dynamic Host Configuration

Protocol (BOOTP/DHCP) 67

Chapter 6: TRANSPORT LAYER PROTOCOLS

193

SOCKETS

Chapter 6: TRANSPORT LAYER PROTOCOLS

194

NOVELL SPX AND NCP

Novells NetWare operating system has two

connection-oriented protocols that function at the transport layer:

SPX

NCP

Chapter 6: TRANSPORT LAYER PROTOCOLS

195

SPX Characteristics
SPX is the acronym for Sequenced Packet Exchange. SPX is a connection-oriented protocol.

It provides packet acknowledgment and flow control.

It is used infrequently by NetWare. Messages are carried in Internet Packet Exchange (IPX)

datagrams.

Chapter 6: TRANSPORT LAYER PROTOCOLS

196

NCP Characteristics
NCP is the acronym for NetWare Core Protocol. NCP is used for NetWare file sharing traffic. It is much more frequently used than SPX. Messages are carried in IPX datagrams. NCP requires an acknowledgment for each

transmitted message.

Chapter 6: TRANSPORT LAYER PROTOCOLS

200

SUMMARY
Connection-oriented transport layer protocols like

TCP, SPX, and NCP provide guaranteed, reliable delivery of datagrams.

They all exhibit the same characteristics: sequencing,

acknowledgments, flow control, error correction and detection, session establishment, and teardown.

Connectionless transport layer protocols like UDP

provide fast but unreliable delivery of datagrams.

control, or error correction. There is no session establishment or teardown.

They do not use sequencing, acknowledgments, flow

201

Chapter 7

TCP/IP

Chapter 7: TCP/IP

202

TCP/IP History
Developed in the 1970s Created for use on the ARPANET Used by UNIX Predates the PC, the Open Systems

Interconnection (OSI) model, and Ethernet

Platform and operating system independent

Chapter 7: TCP/IP

203

TCP/IP Standards
Developed using a collaborative process Published as Requests for Comments (RFCs) by

the Internet Engineering Task Force (IETF)

In the public domain

Chapter 7: TCP/IP

204

Advantages of a Multilayered Design

Platform independence Separate protocols make it

easier to support a variety of communicating Platforms

Quality of service Provide level of service required

Simultaneous development Can develop various

protocols simultaneously

Chapter 7: TCP/IP

205

TCP/IP AND THE OSI MODEL

Peer-to-Peer Communication

Chapter 7: TCP/IP

207

THE LINK LAYER

Transmission Control Protocol/Internet Protocol

(TCP/IP) link layer protocols include

Serial Line Internet Protocol (SLIP) Point-to-Point Protocol (PPP)

Chapter 7: TCP/IP

208

THE INTERNET LAYER

The TCP/IP internet layer is equivalent to the Open

Systems Interconnection (OSI) network layer.

IP Address Resolution Protocol (ARP) Internet Control Message Protocol (ICMP) Routing protocols:

Examples of Internet layer protocols include

Routing Information Protocol (RIP) version 1 and 2 Open Shortest Path First (OSPF)

Chapter 7: TCP/IP

209

THE TRANSPORT LAYER

The TCP/IP transport layer is equivalent to the OSI

transport layer.
TCP UDP

Examples of transport layer protocols include

Chapter 7: TCP/IP

210

THE APPLICATION LAYER

The TCP/IP application layer is equivalent to the session,

presentation, and application layers in the OSI model.

Domain Name System (DNS) Dynamic Host Configuration Protocol (DHCP)

Examples of application layer protocols include

File Transfer Protocol/Trivial File Transfer Protocol (FTP/TFTP) Hypertext Transfer Protocol (HTTP) Simple Mail Transfer Protocol (SMTP) Telnet Simple Network Management Protocol (SNMP)

Chapter 7: TCP/IP

211

ARP
The Address Resolution Protocol (ARP) resolves

logical network layer addresses to Media Access Control (MAC) addresses.

ARP is defined in Request for Comments (RFC) 826.

ARP requests and replies are broadcasts that can

be generated by end systems and routers. routers.

ARP broadcast messages are not forwarded by

THE INTERNET CONTROL MESSAGE PROTOCOL (ICMP)

The ICMP protocol (described in RFC 792) is a

Chapter 7: TCP/IP

217

connectionless network layer messaging protocol.

and routers:

Two types of messages can be sent by end systems

Error and diagnostic. Used to report error conditions

and perform diagnostic tests on a network system

Query. Used to request information from another

Chapter 7: TCP/IP

222

IP ROUTING BASICS
Routers are network layer(Layer 3) devices that
Connect similar or dissimilar data-link layer architectures to

form an internetwork.

Use route tables to forward datagrams across an internetwork.

Datagrams are forwarded based on the logical destination

network layer address. The best path selection is determined by the least cost metric. Routes to remote destinations are learned in two ways:
Statically
Dynamically

Chapter 7: TCP/IP

223

DIRECTLY CONNECTED NETWORKS

Chapter 7: TCP/IP

224

STATIC ROUTES
Static routes are manually configured by an

administrator. network.

There must be one static route for each destination

There must be a default static route. When a specific route becomes unavailable, a new

static route must be added and the old one must be removed.

Static routes do not generate broadcast traffic.

Suitable only for small networks

Chapter 7: TCP/IP

225

DYNAMIC ROUTES
Dynamic routes are automatically learned and

advertised by routing protocols.

Routers use dynamic routing protocols to build their

route tables and advertise route information. very quickly to changes in the network.

Routing protocols, such as RIP and OSPF, can adapt

Routes are either broadcast or multicast.

The best path selection is based on metrics.

Chapter 7: TCP/IP

226

DYNAMIC ROUTES (CONT.)

Suitable for large networks Automatically compensates for network

infrastructure changes

Reduces administrative workload

Chapter 7: TCP/IP

227

ROUTING TABLE INFORMATION

Each route entry includes the following information:
The destination network and subnet mask The IP address of the next gateway (or router) used to

reach the destination destination

The specific outgoing interface used to reach the

The metric value associated with the route

Chapter 7: TCP/IP

228

MICROSOFT WINDOWS XP ROUTING TABLE

Chapter 7: TCP/IP

229

STATIC ROUTE EXAMPLE

Chapter 7: TCP/IP

230

STATIC ROUTE COMMANDS

The configuration of a static route varies, depending

on the operating system of the computer or router you are using.

either

For Microsoft Windows Server 2003 you can use

Route.exe command line interface Routing And Remote Access Console

Chapter 7: TCP/IP

231

STATIC ROUTE COMMANDS (CONT.)

For UNIX and Linux systems, use the Route

command line interface.

For NetWare servers, you can use either

Routecon.nlm or Inetcfg.nlm.

Chapter 7: TCP/IP

232

ROUTE.EXE SYNTAX
To add a static route, use the following command

line syntax:

ROUTE ADD [destination network] MASK [subnet

mask] [local interface address] IF [local interface number] METRIC [metric value for route]

For example:
ROUTE ADD 192.168.3.0 MASK 255.255.255.0

192.168.2.2 IF 1 METRIC 1

Chapter 7: TCP/IP

233

DYNAMIC ROUTING PROTOCOLS

Routers use dynamic routing protocols to advertise

and learn about networks.

Distance vector Link state

There are two types of routing protocols:

Chapter 7: TCP/IP

237

RIP (Routing Information protocol)

RIP is a distance vector routing protocol. There are two versions of RIP:
RIP version 1, or RIP v1 (defined in RFC 1058) RIP version 2, or RIP v2 (defined in RFC 2453)

RIP uses the least number of hops to determine the

best path to a destination. unreachable).

The maximum hop count is 15 (16 = destination

Chapter 7: TCP/IP

241

OSPF (Open Shortest Path First)

OSPF is a link state routing protocol defined in RFC

2328.

OSPF uses link costs with the lowest values to

determine the best path to a destination. network.

Routers maintain a database of routes for the entire

Routers exchange route information through

multicast advertisements.

OSPF supports load balancing and authentication.

Chapter 7: TCP/IP

242

APPLICATION LAYER PROTOCOLS

Provide the communication between a client Run on TCP or UDP

program and a server program across a network

Chapter 7: TCP/IP

243

DNS (Domain Name System)

Resolves Internet Protocol (IP) host names to logical

network layer addresses (converts IP addresses to Hostnames)

Runs on top of UDP or TCP

Uses well-known port 53

Chapter 7: TCP/IP

244

DNS (Domain Name System) (CONT.)

A hierarchical namespace for computer networks

Identifies computers using names composed of 3 or more words, separated by periods. Common Top Level Domains .edu .gov .mil .com .net .org

Chapter 7: TCP/IP

245

DNS (Domain Name System) (CONT.)

Top Level Domain

Second Level Domain

Chapter 7: TCP/IP

DNS Name Resolution

246

The DNS Name Resolution Process

Chapter 7: TCP/IP

247

DHCP (Dynamic Host Configuration Protocol)

Facilitates the automatic assignment of IP

DHCP

addresses

Runs on top of UDP or TCP Uses well-known server port 67 and client port 68

Chapter 7: TCP/IP

248

FTP (File Transfer Protocol)

Is a connection-oriented file transfer protocol Runs on top of TCP Uses well-known server ports 21 (for control) and 20

(for data)

Chapter 7: TCP/IP

249

TFTP (Trivial File Transfer Protocol)

Is a connectionless file transfer protocol Runs on top of UDP Uses well-known server port 69

Chapter 7: TCP/IP

250

HTTP (Hypertext Transfer Protocol)

Used to access Web services Runs on top of UDP or TCP Uses well-known server port 80

Chapter 7: TCP/IP

251

SMTP (Simple Mail Transfer Protocol)

Is an e-mail protocol Runs on top of TCP Uses well-known server port 25

Chapter 7: TCP/IP

252

SNMP (Simple Network Management Protocol)

Allows SNMP management devices to query clients

for information and set network traps. Use to gather information about the network

Runs on top of UDP or TCP

Uses well-known server port 161

Chapter 7: TCP/IP

253

THE TELNET (Terminal Emulation) PROTOCOL

Is a terminal emulation program that allows remote

access and management of network devices

Runs on top of TCP Uses well-known server port 23

Chapter 7: TCP/IP

254

TCP/IP CONFIGURATION PARAMETERS

Each TCP/IP Windows host must be configured with

the following parameters:

IP Address Subnet Mask

Default Gateway
DNS Server Address Windows Internet Name Service (WINS) Server

Address Network Basic Input/Output System (NetBIOS)/Host Name

Chapter 7: TCP/IP

255

TCP/IP AND WINDOWS

All current versions of Windows use the TCP/IP

protocol stack by default.

When the operating system detects a network

interface adapter, it automatically installs the network interface device driver and the following TCP/IP modules:
Client for Microsoft Networks File and Print Sharing for Microsoft Networks Internet Protocol (TCP/IP)

Chapter 7: TCP/IP

257

INSTALLING TCP/IP COMPONENTS

Chapter 7: TCP/IP

258

THE INTERNET PROTOCOL (TCP/IP) PROPERTIES DIALOG BOX

Chapter 7: TCP/IP

259

THE IP SETTINGS TAB

Chapter 7: TCP/IP

260

THE DNS TAB

Chapter 7: TCP/IP

261

THE WINS TAB

Chapter 7: TCP/IP

262

THE OPTIONS TAB

Chapter 7: TCP/IP

263

SUMMARY
The TCP/IP protocol stack consists of four layers: link,

internet, transport, and application. ARP resolves logical network layer addresses to MAC addresses. ICMP is a messaging protocol used to report IP errors and query hosts for information. Routers connect networks. They use static or dynamic routing protocols to learn and advertise routes. Application layer protocols provide services to IP clients, such as file transfer and e-mail capability. IP hosts must be configured with an IP Address, Subnet Mask, Default Gateway, DNS Server Address, WINS Server Address, and other parameters to communicate on a network.