Tài liệu WAN Technologies pdf

This chapter examines the methods used to transport data over wide area networks.We will cover common WAN technologies, such as Frame Relay and ATM, andhow they are configured on Cisco r

○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○

9

WAN Technologies

Terms you’ll need to understand:

✓ Integrated Services Digital Network (ISDN)

✓ Point-to-Point Protocol (PPP)

✓ Dial on Demand Routing (DDR)

✓ High-Level Data Link Control (HDLC)

✓ X.25

✓ Frame Relay

✓ Leased lines

✓ Asynchronous Transfer Mode (ATM)

Techniques you’ll need to master:

✓ Differentiating among wide area network services,including ISDN, X.25, Frame Relay, ATM, andleased-line protocols

✓ Recognizing key WAN terms for ISDN, Frame Relay, X.25,ATM, and leased-line protocols

✓ Using common commands to view the status ofWAN links

This chapter examines the methods used to transport data over wide area networks.

We will cover common WAN technologies, such as Frame Relay and ATM, andhow they are configured on Cisco routers

Also, this chapter covers the following WAN-related CCIE blueprint objectives,

as laid out by the Cisco Systems CCIE program:

➤ Integrated Services Digital Network (ISDN)—Link Access Procedure on the

D channel (LAPD), Basic Rate Interface (BRI) framing, Primary Rate Interface(PRI) framing, signaling, mapping, dialer map, interface types, B/D channels,PPP Multilink

➤ X.25—Addressing, routing, Link Access Procedure Balanced (LAPB), error

control/recovery, windowing, signaling, mapping, Switched Virtual Circuit(SVC)/Permanent Virtual Circuit (PVC), Protocol Translation

➤ Frame Relay—Local Management Interface (LMI), Data Link Connection

Identifier (DLCI), permanent virtual circuit (PVC), framing, traffic shaping,Forward Explicit Congestion Notification (FECN), Backward ExplicitCongestion Notification (BECN), Committed Information Rate (CIR),Discard Eligibility (DE), mapping, compression

➤ Asynchronous Transfer Mode (ATM)—SVC/PVC, ATM Adaptation Layer

(AAL), Service Specific Connection Oriented Protocol (SSCOP), Network Interface (UNI), Network-Network Interface (NNI), Interim LocalManagement Interface (ILMI), cell format, quality of service (QoS), RFC

User-1483, RFC 1577, Private Network-Network Interface (PNNI), Interswitch Signaling Protocol (IISP), mapping

Interim-➤ Leased-line protocols—High-Level Data Link Control (HDLC),

Point-To-Point Protocol (PPP), async, modems, compression

➤ Dial on Demand Routing (DDR)—Short term WAN connections, dial backup

➤ Physical layer—Synchronization, Synchronous Optical Network (SONET),

on the following topics:

Due the to high cost of WANs, Cisco introduced Dial on Demand Routing (DDR)

to provide a WAN connectivity only when end user data is sent across to a remote

or central location DDR is used to reduce the ongoing costs of having a dedicatedWAN circuit active all the time

We will begin our discussion of WAN technologies by exploring Integrated vices Digital Network (ISDN) first

Ser-ISDN

Integrated Services Digital Network (ISDN) is a digital service that enablesnetwork users to send and receive data, voice, and video transmissions over anetwork ISDN offers a variety of link speeds, ranging from 64Kbps to2.048Mbps Therefore, many small- and medium-sized companies find thatISDN is a viable network solution

Basic Rate and Primary Rate Interfaces

ISDN can be supplied by a carrier in two main forms—Basic Rate Interface(BRI) and Primary Rate Interface (PRI) An ISDN BRI consists of two 64Kbpsservices (B channels) and one 16Kbps signaling channel (D channel) An ISDNPRI consists of 23 B or 30 B channels, depending on the country In NorthAmerica and Japan, a PRI service consists of 23 B channels In Europe and Aus-tralia, a PRI service consists of 30 B channels A signaling channel (or D chan-nel) is used in a PRI service and is a dedicated 64Kbps channel

Note: The effective throughput of a PRI service with 23 channels is 1.472Mbps

(23x64Kbps) With 30 B channels, the effective throughput is 1.920Mbps

(30x64Kbps) The International Telecommunications Union (ITU) defines the

standards for ISDN.

ISDN Framing and Frame Format

The ISDN Physical layer provides the ability to send outbound traffic andreceive inbound traffic by transmitting binary bits over the physical media TheISDN Data Link layer provides signaling, which ensures that data is sent andreceived correctly

The signaling protocol used in ISDN is called the Link Access Procedure on the D channel (LAPD) Figure 9.1 shows LAPD’s frame format.

The flag field in an LAPD frame (see Figure 9.1) is used to delimit the frame.The frame control status field is a calculation used to determine whether anyerrors have occurred during a frame’s delivery If the error-checking calculationresult differs from the value in the original frame, an error is generated by thenetwork or router and sent to higher layers for retransmission

Note: The LAPD frame format is similar to the High-Level Data Link Control

(HDLC), which is discussed later in this chapter HDLC is used for point-to-point connections.

Layer 3, the Network layer, of the ISDN model is used to provide call establishment,call termination, and information transfer Before we look at an example ofconfiguring ISDN on a Cisco router using a simple network design, you need

to understand leased line protocols so you can apply this knowledge to complexscenarios, such as authentication techniques in Point-to-Point protocols (PPP)

Leased Line Protocols

A leased line is a service provided by a carrier that maintains a connection betweentwo remote networks separated by some geographical region These remote sitescan range from SOHO (small office home office) to satellite corporate offices

1 Control

1 FCS

1 Flag

Variable Length Data

SAPI C/R EA TEI EA

SAPI – Service Access Point Identifier

C/R – Command/Response

EA – Extended Access

TEI – Terminal Endpoint Identifier (all ones indicate a broadcast)

Field length in bytes

Figure 9.1 LAPD frame format.

HDLC is used in a leased-line setup HDLC is a connectionless protocol that

relies on upper layers to recover any frames that have encountered errors across aWAN link Because HDLC is proprietary, you cannot use HDLC between aCisco router and another vendor HDLC is the default encapsulation on Ciscoserial interfaces

Cisco routers use HDLC encapsulation, which is proprietary This proprietarynature is due to the fact that Cisco added an address field in the HDLC frame,which is not present in the HDLC standard, this field is used by Cisco devices toindicate the type of payload (protocol) Cisco routers use the address field in anHDLC frame to indicate a payload type, but other routers or manufacturers thatimplement the HDLC standard do not use the address field

Point-to-Point Protocol (PPP)

PPP was designed to transport user information between two WAN devices

(also referred to as point-to-point links) PPP was designed as an improvement

over Serial Line Internet Protocol (SLIP) SLIP provided basic IP connectivity.When PPP encapsulation is configured on a Cisco WAN interface, the networkadministrator can carry protocols such as IP and IPX as well as many others.Cisco routers support PPP over asynchronous lines, High-Speed Serial Interfaces(HSSI), ISDN lines, and synchronous serial ports PPP has the added function

of allowing authentication to take place before any end user data is sent acrossthe link

The following three phases occur in any PPP session:

1 Link Establishment—Link Control Program (LCP) packets are sent to

con-figure and test the link

2 Authentication (optional)—After the link is established, authentication can

be used to ensure that link security is maintained

3 Network layers—In this phase; Network Control Program (NCP) packets

deter-mine which protocols will be used across the PPP link An interesting aspect

of PPP is that each protocol (IP, IPX, and so on) supported in this phase isdocumented in a separate RFC that discusses how it operates over PPP

LCP

Link Control Protocol (LCP) is used to establish, configure, and test the linkbetween two devices, such as Cisco routers LCP provides the necessary negotiationsbetween end devices to activate the link Once the link is activated, but no data

is flowing, the next phase of the PPP session can take place, authentication (ifconfigured) and the NCP

technique is known as shared secrets in that both devices know the secret (password),

but they never talk about it directly PAP passwords are sent in clear text, whichmeans they are clearly visible on the wire

NCP

PPP uses Network Control Program (NCP) packets to allow multiple protocoltypes to transfer across WANs from point to point IP Control Program (IPCP)allows IP connectivity, and IPXCP allows IPX connectivity

Figure 9.2 displays the PPP model compared to the OSI model

PPP consists of three main components:

➤ High-Level Data Link Control (HDLC)—Provides for encapsulating datagrams

over PPP links

➤ Link Control Protocol (LCP)—Establishes, configures, and tests a PPP link.

➤ Network Control Program (NCP)—Configures many different network layer

Figure 9.2 The PPP model.

Configuring ISDN on a Cisco Router

Configuring ISDN on a Cisco router requires two main steps:

1 Identify the ISDN switch type in global configuration mode

2 Configure the desired interface parameters

Figure 9.3 shows two Cisco routers connected via an ISDN BRI service On

Cisco routers, a BRI service is indicated by the interface name of BRI For PRI services, the router identifies the service with either T1 (23 B channels) or E1

(For 30 B channels) In this example, router R1 will be configured using the BRIport on the Cisco router to send user data as required across one B channel, andthe ISDN switch type is basic-net3 We will also be using CHAP authentication.Listing 9.1 (which is truncated) displays the configuration for router R1

Note: The BRI interface name is a combination of BRI along with the BRI interface

number (for example, BRI1, BRI2, and so on).

Listing 9.1 The show running-confiig command on R1.

R1#show running-config

version 12.0

hostname R1

enable password 7 1511021F0725

username R2 password cisco

! Define the isdn switch type with the following command

isdn switch-type basic-net3

! Permit all IP traffic across B channel

dialer-list 1 protocol ip permit

!

line con 0

line vty 0 4

login

In the configuration shown in Listing 9.1, the Point-to-Point Protocol is used to

run over the ISDN B-channel The dialer-group 1 command is associated with the dialer-list command, which identifies that all IP traffic across the link This is also

known as defining interesting traffic Interesting traffic is data that is important enough

to the end user to warrant bringing up the WAN connection The IOS command

dialer map ip maps the next hop address to router R2 to the Sydney, Australia, ISDN

phone number 0293353020 The router in Listing 9.1 is configured with the cast routing protocol IP RIP IP RIP sends all updates as IP broadcasts; hence, the

broad-broadcast keyword used in the dialer map statement Using IP RIP or any other

broadcast protocol ensures that the BRI link always remains active Only IP basedtraffic will activate the link, and while the link is active, any other protocol may be

carried across the WAN You use the dialer-list command to advise the router what

protocol, can activate the link For example, if you wanted to enable IPX to activatethe link (or interesting traffic), you would have to add the following command:

dialer-list 1 protocol ipx permit

Listing 9.2 displays a successful ping from router R1 to router R2 after PPP hasbeen configured

Listing 9.2 The ping command on router R1.

R1#ping 131.108.1.2

Type escape sequence to abort.

Sending 5,100-byte ICMP Echos to 131.108.1.2,timeout is 2 seconds:

BRI0 131.108.1.2/24

ISDN Number 0293353020

Figure 9.3 ISDN configuration example.

Table 9.1 displays some useful ISDN-related show and debug IOS commands.

PPP multilink is simply the ability to add more B channels together sothat bandwidth is increased from 64Kbps up to 30 B channels or

1.920Mbps

Now that we have covered ISDN, let’s move onto a protocol that willenable the least expensive method of allowing communication

between two remote sites, namely Dial on Demand Routing (DDR)

Dial on Demand Routing (DDR)

Dial on Demand Routing (DDR) enables routing information to be initiatedand closed as required by transmitting stations, such as PCs A DDR link is onlyoperational when information exchange is required Typically, DDR is deployed

as a backup connectivity solution in case the primary WAN link goes down.DDR is used in environments with a low volume of traffic over an ISDN or

Public Switched Telephone Network (PSTN) DDR can also be used as a backup to

a dedicated leased-line service Figure 9.4 displays a typical situation in whichDDR and dial backup might be used

Figure 9.4 displays two Cisco routers (named Melanie and Ben) connected over anISDN or PSTN line (ISDN would require a BRI interface on the Cisco router, andPSTN would require an external modem) The routers, Melanie or Ben, have onlylow volume traffic to send DDR can be used to maintain a link between to tworouters when required This setup results in financial savings over a long period oftime In general, the process of transferring data between two routers using DDRgoes like this:

1 Traffic defined as interesting arrives at the router forcing the router to

acti-vate the dial up interface in order to transmit the data properly

Table 9.1 ISDN show and debug commands.

IOS Command Description

show isdn interface Displays information on a specific B channel.

show isdn Displays information about memory, layer 2, and

layer 3 timers.

debug bri Displays information about ISDN BRI routing activity.

debug isdn events Displays information about user events that occur on

the interface.

debug isdn q921 Displays layer 2 access procedures.

debug isdn q931 Displays layer 3 information about call setup and call

termination.

2 Before data can be transferred between the two routers, the routers mustactivate the WAN connection and exchange routing information.

3 After routing information is exchanged, data can be transferred

4 After the data transfer is complete, a configurable timeout option expires,and the link disconnects

As mentioned earlier, DDR can also be used to provide a backup option if a mainleased line goes down due to a carrier or router failure Figure 9.4 displays anetwork that contains a primary link and a backup ISDN link

Note: Remember, an ISDN service is tariffed according to usage, so ISDN service is

typically only billed when active.

For illustrative purposes, let’s assume the leased line between the Melanie and Benrouters has failed Either the Melanie or the Ben routers can bring up the ISDNservice to maintain connectivity When the leased line becomes active again, theISDN line can be brought down Listing 9.3 displays a sample configuration on theMelanie router to enable using dial backup

Listing 9.3 Sample DDR backup configuration

ISDN or PSTN

Low volume data

traffic to Router Ben

Low volume data traffic to Router Melanie

Figure 9.4 Typical DDR and dial backup application.

In Listing 9.3, the backup interface bri0 command configures the router to use the BRI0 or ISDN DDR link in case serial0 fails or goes down The backup delay 0 120 command tells the router to wait 0 seconds after the main leased line

has failed and before activating the backup connection and then to wait 120seconds after the main WAN link has been restored before tearing down thebackup link

Another useful purpose of DDR is to provide extra bandwidth for an

existing circuit in periods of high demand The backup load 80 50 IOS

command brings up another WAN circuit if the load on an exiting linereaches 80 percent and brings down the backup link when the existingload reaches 50 percent

X.25

X.25 is an international connection-oriented WAN protocol that was developed

in the 1970s to counter high error rates that were occurring across the physicalmedium These high error rates were due to the fact that truly reliable physicalmedia was not available until the 1980s Early X.25 networks were designed tocompensate for the unreliable analog circuits with built-in features at layer 2 and

3 that provided error detection and correction Because X.25 is orientated, it provides the same features offered by other connection-orientatedservice, namely error control and recovery, windowing, and call setup and calltermination Figure 9.5 shows some of the common elements used in X.25 networks,and Figure 9.6 shows the X.25 model and how it compares to the OSI model

connection-A significant point to remember about X.25 is that X.25 was designed

to operate over unreliable physical media Therefore, it implements

some advanced error checking techniques that allow frames to be

corrected by an X.25 device

PSE or PAD switches X.25 Carrier PAD switches

San Jose

Sydney

DTE DCE

Figure 9.5 X.25 example network.

In Figure 9.5, the Cisco router running X.25 encapsulation on its serial interface

acts as the data terminal equipment (DTE) The carrier provides the X.25 switched backbone or Packet Switching Exchange (PSE), also known as DTE PSEs transfer

data across the carrier network to the end user network The packet assembler/dissembler (PAD) is used to compensate for devices such as PSEs that do notimplement the full functionality of X.25 The protocol running between the dataterminal equipment (DTE) and data communications equipment (DCE) is called

Link Access Procedure Balanced (LAPB) LAPB provides flow control between a

router and an X.25 network, for instance (layer 2 of the X.25 model) The linkbetween the routers shown in Figure 9.5 could be a service that is permanently

active, which is called a permanent virtual circuit (PVC) On the other hand, you

might want to save on costs by configuring a virtual circuit that is only active when

data traverses the link This type of connection is known as a switched virtual circuit

(SVC) In addition to providing a cost savings, SVC connections enable serviceproviders to route around switch failures As you can see in Figure 9.6, the

layer 3 protocol called Packet-Layer Protocol (PLP) provides addressing for

X.25 devices, and the Physical layer is concerned with how bits are transferredacross the physical wire

X.25 Model

Figure 9.6 The X.25 model.

For example, X.21bis is a physical media specification that provides the electricaland mechanical properties required for using X.25 Layer 2 of the X.25 modelprovides LAPB PLP manages the exchanges between the DTE and DCE.

X.25 also has an address called an X.121 address that uniquely identifies eachswitch or host Figure 9.7 shows the format of an X.121 address

As you can see in Figure 9.7, the X.121 address field includes the International Data Number (IDN), which consists of two fields—the Data Network Identification Code (DNIC) and the National Terminal Number (NTN).

Let’s now configure X.25 on a Cisco router in a typical IP environment

Configuring X.25 on a Cisco Router

The steps required to configure a Cisco router for X.25 depend on the ity of the design For a complete guide to configuring X.25, see the “Need toKnow More?” section at the end of this chapter For the simple network shown inFigure 9.8, the following steps are required:

complex-1 Configure an X.25 interface

2 Configure an X.121 address

3 Map any required protocols, such as IP

4 Configure any additional X.25 parameters, such as window sizes

E0 X-121 address is 222

X-121 address is 111

Figure 9.8 X.25 network.

Note: Cisco IOS also permits X.25 protocol translation, as discussed in Chapter 10.

X.25 protocol translation enables you to carry X.25 traffic over an IP cloud, for instance, or enables non-X.25 hosts to communicate to X.25 devices The IOS

command to configure X.25 translation is translate X25 <x.121 address> tcp <ip

address> Chapter 10 provides examples of translation commands, as well as a

typical network situation where translation may be applied.

In the network shown in Figure 9.8, you can see that router R1 has one PVC torouter R2 For this example, assume the X.25 carrier has supplied the X.25address of 111 for R1 and 222 for R2, and IP will run across the X.25 link.Listings 9.4 and 9.5 show the configurations for routers R1 and R2 on theserial WAN connection, which will allow IP

Listing 9.4 Configuring X.25 on router R1

to be running over a broadcast link Therefore, the next hop address must be mapped

using the x25 map ip <next hop address> <remote X.25 address> command The broadcast keyword is used to allow protocols such as RIP and IGRP to send and

received broadcast packets Listing 9.6 shows a successful ping from routers R1

to R2

Listing 9.6 Pinging across an X.25 link

R1#ping 131.108.3.2

Type escape sequence to abort.

Sending 5,100-byte ICMP Echos to 131.108.3.2,timeout 2 seconds:

!!!!!

Table 9.2 shows some common X.25 show and debug commands.

Frame Relay

Frame Relay is a high-speed packet-switching WAN protocol that can supportdata, audio, and video transmissions It was primarily developed in response toimprovements made to networks’ physical medium Frame Relay enables advancederror checking techniques to be left to the higher layers, which means that FrameRelay enables layer 2 (Data Link layer) information to be sent rapidly Contrast thiswith X.25, and you can see how speed advances were gained with Frame Relay

A complete understanding of Frame Relay networks requires you to be familiarwith the technology’s terminology and topology To help summarize FrameRelay networks, Figure 9.9 displays a typical Frame Relay carrier design with anumber of Frame Relay switches set up to provide one or more PVCs betweentwo locations

We will now describe some of the frequently used terms in a Frame Relay:

➤ Committed Information Rate—The rate at which a Frame Relay network agrees

to transfer information under normal conditions

Table 9.2 The X.25 show and debug commands.

IOS Command Description

show x25 map Displays the X.25 maps in use Both dynamic and static

mappings are shown.

show x25 vc Displays X.25 PVCs or SVCs.

show x25 interface Displays X.25 information on interfaces running X.25.

show cdp interface Displays interfaces that are running CDP and their parameters.

debug x25 vc Displays debug information in a virtual circuit setup.

debug x25 interface Assists in troubleshooting X.25 interfaces.

DLCI 250 Frame relay carrier network switches

LMI DLCI 100

LMI

DLCI 100

DLCI 201

DLCI 400

Figure 9.9 Frame Relay terms and topology.

➤ Local Port Speed—The maximum speed at which your local interface on a

router can send information

➤ Committed Burst Rate (B c )—The maximum amount of data that a Frame

Relay internetwork is committed to accept and transmit at the CIR

➤ Excess Burst (B e )—The maximum bits a Frame Relay node will attempt to

transmit after the committed burst rate is exceeded

We will now describe how the Data Link Connection Interface (DLCI) is used

by Frame Relay to identify where a particular frame is sent and how ForwardExplicit Congestion Notification (FECN) and Backward Explicit CongestionNotification (FECN) messages are used to control congestion We will also coverthe Discard Eligibility (DE) bit and how Frame Relay switches manage circuitsusing Local Management Interface (LMI)

Data Link Connection Identifier (DLCI)

The Data Link Connection Identifier (DLCI) in a Frame Relay network defines

a permanent virtual circuit (PVC) or a switched virtual circuit (SVC) DLCIvalues are assigned by the carrier and are only locally significant, because they areused for mappings within a frame switch In the example shown in Figure 9.9,you can see one PVC between the San Jose and Sydney routers Further, the localDLCIs supplied by the carrier are 100 each because each Frame Switch can havethe complete range of DLCIs assigned to it Throughout the carrier cloud, thecarrier determines how a DLCI changes from Frame Switch to Frame Switch Figure 9.10 shows how a DLCI is carried in a Frame Relay packet

1

01111110

Flag

2 to 4 Address Field

Variable Information Field

2 Frame Check Sequence

1 01111110 Flag

DLCI FECN BECN DE EA

DLCI – Data Link Connection Identifier FECN – Forward Explicit Congestion Notification BECN – Backward Explicit Congestion Notification

Another feature of Frame Relay is that it implements some congestion controlmechanisms.

Congestion Control Mechanisms

As you can see in Figure 9.10, the address field contains congestion controlmechanisms used in Frame Relay networks Namely, frame packets can takeadvantage of Forward Explicit Congestion Notification (FECN), BackwardExplicit Congestion Notification (BECN), and Discard Eligibility (DE)

Forward Explicit Congestion Notification (FECN)

Forward Explicit Congestion Notification (FECN) is a flag that is used in FrameRelay networks to control congestion by setting the flag in packets The FECNbit is set in the congestion control field in the Frame Relay packet (as shownearlier in Figure 9.10)

Typically, a Frame Relay service provider will allow a device to burst to a higher rate

than the Committed Information Rate (CIR) The CIR is a rate that is sustainable,

but if the rate is above the CIR rate, FECNs may be used to indicate to a receivingdevice that congestion was encountered along the transmission path coming intothe device (that is, forward) FECNs are typically used by the higher layers of theFrame Relay protocol to slow down data transfer

Frame Relay traffic shaping is an advanced form of congestion control.Special queues can be set up to buffer data and send it in regulatedamounts This results in a mechanism that avoids congestion prob-

lems Traffic shaping is also used in ATM, which is discussed later inthis chapter

Backward Explicit Congestion Notification (BECN)

Backward Explicit Congestion Notification (BECN) is also used in Frame Relaynetworks to control congestion The BECN bit is set by a Frame Relay switchnetwork (remember this device is configured by the carrier) to indicate to thesending device that frames are traveling in the opposite direction of frames thatare encountering a congested path (that is, away from the device receiving them)

A Frame Relay frame with the BECN bit set in the control field (as shown inFigure 9.10) is sent from a switch within the carrier’s network toward a routerthat is sending a large amount of data The router receiving this packet willreduce the rate at which packets are sent out the interface Remember, FrameRelay uses frames, not packets

Discard Eligibility (DE)

The Discard Eligibility (DE) bit is used to indicate which switch frames can bediscarded during periods of high-volume traffic As with FECN and BECN,