Local Area Network can be easy understood since typically every office has a LAN. One could personally see the switches linking assorted computing machine nodes and may be a router to which the switches are connected. Besides we can see the Network Interface Card ( NIC ) at the dorsum of the computing machine where we plug in the overseas telegram with the RJ45 connection. the eye blink light etc. on the card. ( Sometimes when it is non working. we pull out the overseas telegram from the card and stopper it once more to seek our fortune! ! ) All that we know. But how will you set up a connectivity between your Head office located in Chennai and the Branch office in Singapore? What are all the possibilities and whom to near? We know that there is no manner to put a overseas telegram all the manner to Singapore! That will be highly hard and non a good thought in footings of cost and feasibleness besides. So. let’s foremost understand what makes the difference between a LAN and a WAN. The general and immediate reply comes to our head is that LAN is ‘local’ and WAN is a “wide” country web.

The Distance! . But now-a-days we have the radio LAN engineering which can truly cover a ‘wide’ country. Then is it the Bandwidth? Again we see the competition at our door measure that we can acquire gigabit broadbands from assorted service suppliers at much. much cheaper rates. This is non either. What so? By and large. to put up a LAN. we buy computing machines. switches. overseas telegrams. connections. routers etc. But do you purchase any equipment or wire to link your Branch office located at Singapore or Delhi or Bombay? The reply is No. So we own the LAN substructure but we by and large lease WAN substructure from any 3rd party Service Providers or from a Telephone company. WANs by and large carry a assortment of traffic types such as voice. informations and picture. Imagine in your office there are 25 telephone extensions. So you have 25 telephone instruments placed on each officer’s desk wired and you own the substructure.

If you want to name your three Branch offices at Singapore. USA and Canada do you have the substructure? i. e. the wires. telephone instruments connected till your subdivision offices? No. you ‘lease’ them. To do it more clear. you take a telephone line from the telephone company with ISD installation. Now that utilizing your telephone company’s substructure. you can name to Singapore or any other subdivision offices anyplace in the universe! . Similarly we own the LAN substructure and we ‘lease’ the substructure of a 3rd party supplier or a Telco for Wide Area Network ( WAN ) communicating. If you carefully read the old notes on LAN communicating. we came across batch of nomenclatures. Like that here besides we have to larn some nomenclatures. Let us foremost understand the WAN connexion types.

WAN Technolologies Imagine you have a subdivision office in Delhi and another one in Canada. Your demand is that every two proceedingss. you will pass on with each subdivision office. You must hence hold an ever up connectivity for this at the same clip cost effectual. You can travel for a dedicated line ( a synchronal consecutive connectivity ) so that it will be ever on. But to set up a dedicated line. you will hold to pass batch of money. If in instance you are non utilizing the full capableness of the dedicated line and it happens that sometimes you will pass on with your subdivision offices merely for 3-4 hours a twenty-four hours. so passing so much on a dedicated line is non a executable solution. The option is a dial up line. Then every clip you communicate. you have to dial and convey the line up and so direct the information. It is boring and though cost effectual when compared to dedicated line. the bandwidth is besides non guaranteed! . What we need is a connectivity. which is ever ON at the same clip cost effectual besides. One of such services is called Fame Relay. Frame relay is a Technology and non a protocol. We will be discoursing about frame relay in a few proceedingss. Before that let us rapidly see some nomenclatures. Leased Line: Typically. these are referred to as a point-to-point connexion or dedicated connexion.

A leased line is a pre-established WAN communicating from the Customer premises through the ISP’s substructure ( DCE switches ) to the client of the distant site. If cost is no object. it is truly the best pick. Circuit shift: When you hear the term circuit exchanging. think of phone call. Cost is the large advantage in circuit shift. You merely pay for the clip you really use. No information can reassign before an “end-to-end” connexion is established. Circuit exchanging utilizations dial up modems or ISDN and is used for low bandwidth informations transportations. Every clip you communicate. you have to convey the line up and set up connectivity and merely so you will be able to direct the information. This is non ever up connectivity. Packet shift: This is WAN exchanging method that allows you to portion bandwidth with other companies to salvage money. Packet shift can be thought of as a web that is designed to look like a leased line. yet charges you ( and costs ) more like circuit shift. Frame relay and x. 25 are packet exchanging engineerings. Speeds can run from 56kbps to T3 ( 45 mbps ) . Frame Relay: Frame Relay is a package switched engineering. Frame Relay is a information nexus and physical bed specification. This provides high public presentation. Frame Relay is a replacement to X. 25. Frame Relay provides characteristics for dynamic bandwidth allotment and congestion control.

Compiled by A. K. Asokan. ( E-mail: [ electronic mail protected ])

Page Nos. 3 of 16

ISDN: Integrated Services Digital Network ( ISDN ) is a set of digital services that transmit voice and informations over bing phone lines. No demand to alter the telephone overseas telegram. It defines high velocity digital informations transportation over standard phone lines. In dial-up. the modem is change overing the digital information from computing machine and sends as parallel informations through the phone lines. ISDN can offer a cost effectual solution for distant users who need a higher velocity connexion than parallel dial-up links offer. ISDN is besides a good pick as a backup nexus for other types of links such as Frame Relay or a T1 connexion. ( The velocity of the T1 connexion is 1. 544mbps full semidetached house. Full semidetached house means both ways at the same clip informations can be transmitted ) . It is of import to understand what is a back-up line and how and why ISDN is used as a backup line for chartered line. Please refer the undermentioned diagram.

In the above diagram. there is a Head office at Chennai and the subdivision office in Delhi and both the offices are connected via a leased line through RA and RB routers. These offices are besides connected through an ISDN line as a back-up line. Back-up line agencies. originally the communicating is being sent through the leased line which is ever up. Suppose if the leased line is down for some grounds. so the ISDN line will automatically come up and set up the connectivity provided both the routers RA and RB are configured in such a manner. When the leased line comes up once more. the ISDN backup line will automatically disconnected. Thereby there is no downtime in the connectivity. We may even utilize the ISDN line for all communicating. But it is really dearly-won. Another usage of this set up is to direct ‘interesting traffic’ . What is meant by interesting traffic? Suppose if you have a mail waiter at the Head office and you want to pass on with the mail waiter at the subdivision office. so you can direct these SMTP packages entirely through the ISDN line and all other packages like hypertext transfer protocol and file transfer protocol traffic through the leased line. This has to be configured in the router’s consecutive interface. HDLC: High degree Data nexus control ( HDLC ) is a protocol at the informations link bed. HDLC is a point-to-point protocol used on chartered lines. ( Remember the following diagram? we discussed point-to-point protocol and multi entree webs in the old OSI Layer notes. )

High Level Data nexus Control ( HDLC ) is non intented to encapsulate multiple Network Layer protocols across the same nexus. The HDLC heading carries no designation of the type of protocol being carried inside the HDLC encapsulation. Because of this. each seller that uses HDLC has their ain manner of placing the Network bed protocol. This means that each vendor’s HDLC is proprietary for their ain equipment. To do it clear. Cisco’s HDLC is proprietary to Cisco and it is the default encapsulation used by Cisco routers. It will non pass on with any other vendor’s HDLC execution. So if you have merely Cisco routers. how will you link your routers and web to others? If you have Cisco router and a non-cisco router with a consecutive connexion. you must configure PPP or another encapsulation method such as Frame Relay because the HDLC default won’t work.

Point to indicate protocol ( PPP ) is an industry criterion protocol. Because all multiprotocol versions of HDLC are proprietary. PPP can be used to make point to indicate links between different vendors’ equipment. It uses a Network Control Protocol field in the Data nexus heading to place the Network Layer protocol. It allows hallmark and multi-link connexions and can be run over asynchronous and synchronal links. To sum up the above two paragraphs. protocol means ‘rules’ . If Head office router and Branch office router want to interchange communicating. there must be some ‘protocols’ which both the routers must accept. HDLC ( High Level Data Link Control ) and PPP ( Point-to-point protocol ) are two such protocols. However. if both the routers are of the same seller ( i. e. Cisco ) so HDLC is default. If both the routers are non from the same maker. so HDLC can non be used. Here PPP can be used. However. if both the routers are from the same seller and HDLC is the default. still if you prefer. PPP can be used alternatively of HDLC. But one side HDLC and other side PPP will non work.

For illustration. all of us can talk English. That is like HDLC. A default linguistic communication. But I know Gallic and Mr. Neil besides knows French. Then we have the option either to talk in Gallic or in the default linguistic communication English. But if I speak German and Mr. Neil speaks in Spanish. so we will non be able to pass on anything! Puting up these protocols with the router’s consecutive interface utilizing router bids is what is known as ‘encapsulation’ . Another thing is that IP is non the merely routed protocol. there are other webs like ATM. AppleTalk. IPX/SPX etc. apart from IP. ( You may read one time once more the above 3 paragraphs. if you are non certain of some constructs ) . Point to indicate protocol ( PPP ) is a Data nexus bed protocol that can be used over either asynchronous series ( dial-up ) or synchronal consecutive ( ISDN ) media. It uses something called the LCP ( Link command protocol ) to construct and keep informations nexus connexions. Network control protocol ( NCP ) is used to let multiple web bed protocol ( routed protocols – IP. ATM. AppleTalk etc. ) to be used on a PPP connexion. LCP is a method of set uping. configuring. maintaining and ending the point to indicate connexion and NCP is method of set uping and configuring different web bed protocols.

Since HDLC is the default consecutive encapsulation on Cisco consecutive links and it works great. when would you utilize PPP? The basic intent of PPP is to transport bed 3 packages across a Data nexus bed point to indicate nexus. It is non-proprietary. which means that if you don’t have all Cisco routers. PPP would be needed to be encapsulated on your consecutive interfaces. The HDLC encapsulation would non work because it is proprietary to the specific seller. In add-on. we should understand that PPP can encapsulate several bed 3 routed protocols and supply hallmark. dynamic addressing and name back installation etc. This may be the encapsulation solution of pick for you over HDLC. PPP Authentication methods Before set uping connectivity between two routers. it has to be authenticated each other. There are two methods of hallmark that can be used with PPP links i. e. PAP and CHAP. PAP stands for Password Authentication Protocol and CHAP bases for Challenge Handshake Authentication Protocol. It is upto the Network decision maker to choose PAP or CHAP. However CHAP is the preferable protocol as it is a three manner handshaking protocol.

Password Authentication Protocol ( PAP ) The Password Authentication Protocol ( PAP ) is the less secure of the two methods. When the PPP nexus is foremost established the distant node sends back to the arising router the username and watchword until hallmark is acknowledged ( Refer diagram below ) .

In the above diagram. PAP hallmark is described. RA wants to pass on with RB. As both are configured for PPP. RB knows the username and watchword. When RA sends the username and watchword. RB verifies it and so accept/reject it so that communicating is established / non established. If the username and/or watchword are incorrect. so RB rejects the petition to set up connectivity. PAP is non a strong hallmark protocol because the username and watchword are sent across the nexus as ‘clear text’ and hence there is no protection. Challenge Handshake Authentication Protocol ( CHAP ) The Challenge Handshake Authentication Protocol ( CHAP ) is used at the initial start up of a nexus and at periodic medical examinations on the nexus. to do certain the router is still pass oning with the same host. After PPP finishes its initial nexus constitution stage. the local router sends a challenge petition to the distant device. The distant device sends a value calculated utilizing a one manner hash map called MD5. The local router cheques this has value to do certain it matches. If the value doesn’t lucifer. the nexus is instantly terminated. See diagram below.

Initially a challenge package is sent and a response is received for that with the username and watchword. Then the credence / rejection are established. ( Both the routers send the packages to each other to do certain their individuality ) ( Those who work on the router can do certain this by first encapsulating the consecutive interface with ppp fellow and so debugging the same while conveying the interface down and once more conveying it up ) . Frame Relay Frame Relay is a bit more complex than the leased line networks we have discussed above i. e. HDLC and PPP protocols. These leased line webs are easy to gestate. Not so with Frame Relay. It can be significantly more complex. which is why it is frequently represented as a cloud in networking artworks. I have made every effort to do you understand the constructs though merely certain topologies are explained in item down the page. Please have the diagram referred frequently while you read the accounts so that it will assist you to understand it in a better manner. At some topographic points. there will be repeat of constructs which is intentionally made in order to give an impact for understanding. Let us seek to understand Frame Relay. Read on……

Frame Relay has become one of the most popular WAN services deployed over the past decennary. It is a package switched engineering. Frame Relay engineering saves money over options. By default it is classified as a non-broadcast multi entree ( NBMA ) web. which means that it does non direct any broadcasts. such as RIP updates across the web by default.

In the above diagram. there is a Head office located in Chennai and there is a Branch office located in Delhi. In order to pass on with the two offices. both of them take a frame relay connectivity. The substructure of the service supplier will give you a Virtual Circuit ( VC ) which is ever up and running. Equally long as you pay the charges. it is lasting and hence it is known as Permanent Virtual Circuit ( PVC ) . There will be 100s of PVCs in a frame relay cloud. Frame relay comes in between a leased line and a dial-up line. Under frame relay. the Service Provider commits that any given clip. you are guaranteed with a certain bandwidth. This is known as Committed Information Rate ( CIR ) . You can purchase your CIR. Frame relay is a engineering and non a protocol. Imagine you get a connexion that looked like a leased line and acted like a leased line but allowed you to pay for whatever part of that leased line you really used. That is basically what package switched webs do. You pay for whatever you use.

Frame Relay provides a package switched web to many different clients at the same clip. It is based on the premise that all clients will ne’er necessitate to convey changeless informations all at the same clip. An illustration to throw more visible radiation on the “assumption” could be that in a bank. there are figure of clients who deposited money. The bank can deviate the money for some other investing intents on the premise that all the clients will non come to retreat the money at the same clip. Like that. the Frame Relay connectivity is provided to you on the premise that all the clients will non direct informations invariably so that some part of the bandwidth will ever liberate and can be used by other clients. Frame Relay is usually implemented with criterion unshielded distorted brace ( UTP ) telegraphing from the Service supplier. Devicess on the endorser premises are called client premises equipment ( CPE ) . This is once more connected to a device known as Channel Service Unit / Data Service unit ( CSU/DSU ) and so connected to your company’s router with the consecutive interface. The CSU/DSU ( Modem ) by and large takes the original signal from the supplier and converts its electrical belongingss into the consecutive format for your router. It besides sometimes move as a channel bank. or multiplexer. which breaks the channelized connexion apart so that the voice and informations can be separated. The undermentioned diagram explains the cabling of frame relay from your premises till the service supplier.

Frame relay plants by supplying a part of dedicated bandwidth to each user and besides leting the user to transcend their guaranteed bandwidth if resources on the Telco web are available. The suppliers allow their clients to purchase a lower sum of bandwidth than what they truly use. This is possible because of multiplexing. There are two types of multiplexing. 1. Frequency Division Multiplexing ( FDM ) 2. Statistical Division Multiplexing ( SDM ) In frame relay. we use statistical division multiplexing. This means that if all the clients are conveying at the same clip. so you are assured with the CIR ( assured bandwidth ) . Depends on some clients are non conveying. you get a better bandwidth to that extent. Anything beyond your committed information rate ( CIR ) is a fillip for you and it is known as ‘burst’ . The undermentioned nomenclatures are of import. Access rate and Committed Information Rate ( CIR ) The maximal velocity at which the frame relay interface can convey is known as Access rate. CIR the maximal bandwidth of informations guaranteed to be delivered. The clients can repair the CIR it can be 256kbps or 512 kbps etc. depends on their demand. However. in world this is the mean sum that the service supplier will let you to convey.

If these values are the same. ( entree rate and the CIR ) so the FR connexion is reasonably much like a leased line. However. they can besides be different values. Let us state that you buy an entree rate of T1 ( 1. 54mbps ) and a CIR of 256 kbps. By making this. the first 256kbps of traffic will ever be available to you and it is guaranteed to be delivered. Anything beyond that is called a ‘burst’ which is the transmittal that exceeds your guaranteed 256kbps. and can be any sum upto the T1 entree rate ( if that sum is in your understanding ) . The bringing of the 256 kbps is guaranteed. The “burst” will be delivered on something called best attempt bringing. Or may be non delivered. If your Telco’s equipment doesn’t have the capacity to present at the clip you transmitted. so your frames will be discarded and the DTE ( your router ) will be notified so that it will retransmit the same.

Timing is everything – you can shout informations out at six times your guaranteed rate of 256kbps ( T1 ) merely if your Telephone company has the capacity available on their equipment at that minute! We have antecedently discussed that multi-access webs require designation ( LAN Communication notes ) and point-to-point communicating doesn’t need designation of finish. Frame relay is multi-access in nature. But it will non air so we can state non-broadcast multi-access ( NBMA ) . ( retrieve the illustration of a telephone exchange? It will non direct a call to all the lines it is connected to ; but it can entree multiple finishs. In other words. it will non ‘broadcast’ but multi-access is possible. ) So frame relay is an NBMA web and since it is multi entree. designation of the finish is necessary. Non-broadcast multi entree web ( NBMA ) means that it does non direct any broadcasts like RIP updates ( periodic updates ) across the web by default. Let us now see what is the placing mechanism by frame relay. If you remember. we besides have discussed antecedently that Ethernet on a LAN environment plants based on MAC reference.

Similarly frame relay is based on a figure known as DLCI Numberss ( Delsi Numberss ) . DLCI stands for Data Link Connection Identifier. ( Technically. unlike MAC reference. DLCIs do non stipulate the physical port. They specify the logical nexus between two systems. It is merely practical and non physical ) LAN – MAC reference ( Media Access Control ) WAN ( FR ) – DLCI ( Data Link Connection Identifier ) Every VC ( practical circuit ) is identified by DLCI Numberss. The Telco gives you your DLCI Numberss. DLCI 100. DLCI 150. DLCI 200 etc. DLCI has merely local significance. The Telco’s frame relay switch maintains the DLCI Numberss. There can non be two indistinguishable Numberss in a switch. If you need to name Singapore office. you need a telephone connexion with ISD installation which you may acquire from a Telco. Like that. in order to put up web communicating with Singapore office. you must acquire a frame relay connexion with a committed information rate ( CIR ) ( your required bandwidth ) from the Service Providers. Mention the undermentioned diagram.

In the above diagram. router RA is in Chennai and router RB is in Singapore. Both the offices are connected via a frame relay connectivity. Mention the cloud. it has an ISP switch. which provides a PVC ( Permanent virtual circuit ) till the other side ISP which provided a frame relay connectivity to the Singapore office. The PVC has DLCI Numberss 102 for Chennai office and 201 for Singapore office. ( Think of the PVC as a practical pipe ) When RA wants to direct a package to RB it checks its routing table whether there is any path specified to RB whose finish IP reference is 20. 0. 0. 2. See the routing tabular array of RA which says that any package you want to direct to IP reference 20. 0. 0. 2. send it to Serial 0 ( S0 ) interface of the same router ( IP address 20. 0. 0. 1 ) . Accordingly it sends the package to S0 interface. Since the consecutive 0 interface of RA is encapsulated with frame relay. it has another tabular array called frame relay MAP tabular array. Hence it looks into the Frame relay map tabular array as to what to make with the package whose finish IP reference is 20. 0. 0. 2.

The frame relay map tabular array Tells yaar. for 20. 0. 0. 2. you put the package into DLCI figure 102. It puts the package into DLCI figure 102. Since the line is ever up. it reaches the Telco’s frame relay switch and the switch. switches the package to DLCI figure 201 of the other side ISP through the frame relay cloud. who in bend delivers the package to IP reference 20. 0. 0. 2 which is RB router. Similarly. when RB router wants to direct a answer package. it sends the same to the routers consecutive interface and since the interface is encapsulated by frame relay. it has a celebrity relay map tabular array which tells. to set the package into DLCI figure 201 and it reaches Chennai DLCI 102 and delivers the same to the Chennai router 20. 0. 0. 1 ( please understand that this is the manner it works and in order to understand to the full. we need to understand the LMI type and other constellation inside informations which is omitted in this note. )

Mention the above diagram. in the frame relay cloud. there are 100s of frame relay switches. Frame relay is a synchronal communicating means it synchronises with the clock velocity. The clocking is provided by the informations communicating equipment ( DCE ) which is present in the Telco’s substructure. Frame relay specifies how to run in the local expression which is non to the full described here and it doesn’t stipulate how the frames are traversing the cloud. Mention the undermentioned diagram to understand the cloud a little more.

The Frame Relay ‘cloud’ contains 100s of Frame Relay switches. There are two types of devices exist. The DCE and DTE. DCE stands for Data communications Equipment and DTE is for Data Terminal Equipment. The DCEs are the frame switches in the cloud. And the DTEs are the routers. The DCEs are capable of managing batch of Virtual Circuits ( VCs ) at the same time. Apart from that the DCEs provide “clocking signal” to the DTEs. Clocking signal is needed because Frame Relay is a synchronal protocol and the frames are ‘synchronized with the timing signal’ . There is no start spot and halt spot needed in this instance. Hence it is more efficient and faster. For those who find it is hard to understand please presume that your Telco will give you a frame relay connectivity for informations communicating ( as a telephone connectivity for voice communicating ) . and you must necessitate some equipment like the router and Modem ( CSU/DSU ) unit.

CSU is Channel Service Unit and DSU is Data Service Unit to put up the connectivity. For a telephone call to land in Singapore office. the Singapore office besides must hold equipped with a telephone connectivity from any Telco available in Singapore. As you don’t bother about how the Telcos communicate with each other. the same scenario exist here as to how the Telco does the information communicating through frame relay engineering. in a user position. The thought here is to understand that to pass on with a distant topographic point which is geographically and administratively doesn’t have any boundaries. and to put up a WAN connectivity. we have certain engineerings and one of such engineerings is Frame relay.

There are assorted types of frame relay connectivity. 1 ) Full mesh topology 2 ) Partial mesh topology. and 3 ) Hub and spoke topology The undermentioned diagram explains the full mesh topology.

In the above diagram. the Head office ( HO ) is connected to 5 subdivision offices. All the first subdivision offices are connected each other. This is known as full mesh topology and it provided batch of redundancy and mistake tolerance. Even if one nexus fails. the connectivity is non wholly disconnected. However. expression at the diagram how many PVCs are required to purchase? ! It will be extortionately! . You can hold partial mesh besides which means may be branch office 3 and 5 entirely can be to the full meshed. The option is Hub and spoke topology. Mention the undermentioned diagram.

In the above diagram. the Head office ( HO ) is connected to 5 subdivision offices. However. all the five subdivision offices are connected merely to Head office and non each other. So we need to acquire merely five PVCs. It is less dearly-won to that extent. The lone job with hub and spoke topology is that. in instance the router at the HO is failed. so none of the subdivision offices can pass on each other. Otherwise. it is an first-class manner to link distant locations. The hub and spoke topology is a simple and by and large the easiest to put up. In a hub and spoke topology. every device is connected to a cardinal hub device. ( a router ) which performs the routing work. In the undermentioned diagram a hub and spoke topology is described. See that if the subdivision offices have to reach each other. they have to link through the Head office.

In the above diagram one new construct we are traveling to discourse before we close. Take a close expression at the router RA. It has merely one consecutive interface. But it has been connected to three subdivision offices located in Singapore. USA and Canada severally. All the three subdivision offices are ( encapsulated with frame relay ) connected to a individual consecutive interface. This is possible due to a new construct known as “sub-interfaces? . What are sub-interfaces? Think of a sub-interface as a logical interface. Several subinterfaces will portion a individual hardware interface ( physical interface ) which is called multiplexing.

You can specify sub-interfaces in the consecutive interface of the router. Before making sub-interfaces. the physical consecutive interface must be encapsulated and so you can specify sub-interfaces. By and large one sub-interface per PVC will be created. In our above diagram. there are three sub-interfaces. Each sub-interface must be given single IP references. See the above diagram. the sub-interfaces are given ip references viz. 172. 16. 1. 1. 172. 16. 2. 1 and 172. 16. 3. 1. Sub-interfaces are of two types. 1 ) Point-topoint and 2 ) multipoint. It is really really of import to observe that when you create subinterfaces. the physical consecutive interface must non be given an IP reference but it has to be encapsulated with Frame Relay. Dear Friends. the WAN constructs described above is merely to the point. You are requested and encouraged to read batch of stuffs from standard publications. diaries and periodicals in order to acquire good apprehension of the constructs. Hope this note was utile to you.