Frame Relay

I have confusion regarding frame Relay and leased lines. Suppose we have three branches and one headquater HQ. If we were to use leased lines we would use there T1 lines.

In second case if we use frame relay each branch router DTE connect to switch of frame Relay DCE via TI ( is it true that DTE --DCE LMI uses TI even we use FR )and similarly one T1 is used at HQ to service provider DCE.

How much saving we are going to get in terms of leased lines T1 and CSU/DSU when we use FR and please can you tell me how we are going to use CDU/DSU when we use FR ?.

Find more posts tagged with

Comments

kryolla

layer 1 is T1 and layer 2 is frame relay.

Plazma

T1 = Layer 1 because it deals with the type of signaling on the physical wire.

Frame Relay = Layer 2 because it deals with creating virtual circuits involving Layer 2 addressing and uses DLCI's which are also Layer 2

tech-airman

Manglian,

Manglian wrote:

I have confusion regarding frame Relay and leased lines. Suppose we have three branches and one headquater HQ. If we were to use leased lines we would use there T1 lines.

The International Telecommunications Union (ITU) has designated DS1 as basically "many DS0 channels." The North American implementation of DS1 is T-1. The European implementation of DS1 is E-1. Since your location is listed as London, your leased line would be an E-1 and not a T-1. So I have to ask which of the three branch(es) and/or HQ is/are located in London?

Manglian wrote:

In second case if we use frame relay each branch router DTE connect to switch of frame Relay DCE via TI ( is it true that DTE --DCE LMI uses TI even we use FR )and similarly one T1 is used at HQ to service provider DCE.

Frame relay is a different WAN technology compared to leased line (e.g. T-1). You mentioned "...frame Relay DCE via T1..." which is inaccurate. You can have either a T-1 link running at 1.544 Mbps or a Frame Relay link running at 1.544 Mbps Commited Information Rate or CIR for short not both on the same link.

Manglian wrote:

How much saving we are going to get in terms of leased lines T1 and CSU/DSU when we use FR and please can you tell me how we are going to use CDU/DSU when we use FR ?.

Only you can calculate cost savings based on the cost of services in your area. However, from a technological cost savings, for leased line, you'd need to upgrade the HQ router with another serial port for each leased line link however using frame relay with the HQ requires a single router with a single serial port at HQ instead of more than one serial port.

miller811

The main benefit of using Frame Relay would be you would only need a single connection at the host location (HQ).
If you were using just T1 connections, and had three remote locations you would need 6 DSU's and would be paying for three dedicated connections, 1 for each remote location.

Manglian

Tech-airman

Thanks for your reply

Frame relay is a different WAN technology compared to leased line (e.g. T-1). You mentioned "...frame Relay DCE via T1..." which is inaccurate. You can have either a T-1 link running at 1.544 Mbps or a Frame Relay link running at 1.544 Mbps Commited Information Rate or CIR for short not both on the same link

The bottom line is that you need some physical connectivity to get from your
equipment to the frame switch. A T1 is a option ( depent upon physical location of the network ). The T1 will
go from your site to the frame switch (not to your other branch).

I mean how you going to connect your router at your site with frame relay switch ( not to other branch) ?

I used three branch , one HQ example from Todd lammle book 6th edition page 801

tech-airman

Manglian wrote:

Tech-airman

Thanks for your reply

Frame relay is a different WAN technology compared to leased line (e.g. T-1). You mentioned "...frame Relay DCE via T1..." which is inaccurate. You can have either a T-1 link running at 1.544 Mbps or a Frame Relay link running at 1.544 Mbps Commited Information Rate or CIR for short not both on the same link

The bottom line is that you need some physical connectivity to get from your
equipment to the frame switch. A T1 is a option ( depent upon physical location of the network ). The T1 will
go from your site to the frame switch (not to your other branch).

I mean how you going to connect your router at your site with frame relay switch ( not to other branch) ?

I used three branch , one HQ example from Todd lammle book 6th edition page 801

Manglian,

It seems like your confusion seems to be coming from your attempt at trying to squeeze two separate concepts into a single understanding. You need to understand each technology separately. In the case of leased line, you can use a T-1, E-1, or whatever national telecommunications implementation of the ITU DS1 standard from end to end, in other words, from site to site. In a leased line, there are intermediate telco switches involved NOT frame switches. The thing about leased line is that you as the organization is the only user of the WAN link and in the case of T-1, the entire 24 DS0 channels are available to you for traffic.

On the other hand, if you choose frame relay as your WAN link type, then your site's router connects to a frame switch owned by the Service Provider. So to answer your question of "I mean how you going to connect your router at your site with frame relay switch ( not to other branch) ?" completely depends on your local Network Service Provider so you'll have to ask them. However, in order to get an idea how frame relay works, find out how to configure and use a router to create a frame relay switch. As far as actual implementation of frame relay within the "last mile" for where you are completely depends on your Service Provider.

Manglian

Thanks Tech-airman

It seems like your confusion seems to be coming from your attempt at trying to squeeze two separate concepts into a single understanding. You need to understand each technology separately

I understand leased lines and FR are two different WAN technlogies.

Can you please explain me the following paragraphs from Todd lammle book.

suppose you have a central HQ with three branch offices.If you were to connect each branch office to HQ using T1 , you would need three serial interfaces on your router at HQ, one for each T1. Suppose you use Frame Relay PVCs instead. You could have a T1 at each branch connected to a service provider and only a siingle TI at HQ

There would be three PVCs on the single TI at HQ one going to each branch .And even though there's only a single interface and a single CSU/DSU, the three PVCs function as three separate circuits.Remember what I said about saving money? How much for two additional T1 interfaces and a pair of CSU/DSUs? Answer : A lot

Many thanks once again

tech-airman

Mangalian,

Manglian wrote:

Can you please explain me the following paragraphs from Todd lammle book.

suppose you have a central HQ with three branch offices.If you were to connect each branch office to HQ using T1 , you would need three serial interfaces on your router at HQ, one for each T1. Suppose you use Frame Relay PVCs instead. You could have a T1 at each branch connected to a service provider and only a siingle TI at HQ

There would be three PVCs on the single TI at HQ one going to each branch .And even though there's only a single interface and a single CSU/DSU, the three PVCs function as three separate circuits.Remember what I said about saving money? How much for two additional T1 interfaces and a pair of CSU/DSUs? Answer : A lot

Many thanks once again

Since I do not have Lammle's book, I'll have to rely on what you've said.

To help you to understand this situation, it helps to look at it from "here's this technology, here's this other technology, and how the second technology is better than the first technology" perspective. With that said, here goes my explanation.

First, we'll have to speak in terms of the non-technical aspect of engineering which in this case are buildings or sites. So let's say you had an organization with a single site which effectively also makes that site the Headquarters or HQ for short. Then, let's say that organization grows to another site which we'll call Branch 1. So in this case, you'll need a router at the HQ site with a single serial port and at Branch 1 you'd need a router with a single serial port. So the WAN network diagram would look like...

Leased line

(HQ)[Serial 0]--------z-----------[Serial 0](Branch1)

Now, let's say another branch opened up, let's call it Branch2. Now it's possible to start comparing between leased line versus frame relay.

Leased Line

    [Serial 0]-------------z-----------[Serial 0](Branch1)
(HQ)
    [Serial 1]-------------z-----------[Serial 0](Branch2)

Frame Relay

                          +-------z----------[Serial 0](Branch1)
                         /
(HQ)[Serial0]============
                         \
                          +-------z-----------[Serial 0](Branch2)

So even under a single HQ and two branch situation, in the leased line case, you'd need to physically upgrade the router to another one with two serial ports instead of one serial port. However, with Frame Relay, you can keep the same single Serial interface router at the HQ site. Now, let's say this organization grew again to a third branch which is the situation in the section of the book you are apparently citing.

Leased Line

    [Serial 0]--------------------z-----------------[Serial 0](Branch1)
(HQ)[Serial 1]------------------z-----------------[Serial 0](Branch2)
    [Serial 2]-------------------z------------------[Serial 0](Branch3)

Frame Relay
                     
                              +-------z---------[Serial 0](Branch1)
                             /
(HQ)[Serial 0]===============--------z----------[Serial 0](Branch2)
                             \
                              +-------z----------[Serial 0](Branch3)

So as you can see from the extension of the two branch to three branch scenario, if you upgraded your HQ router from a one serial port router to a two serial port router, well, you're going to need to upgrade that router again from a two serial port router to a three serial port router, just so you can connect to each of the other branches. However, with the Frame Relay solution, you can keep the same router that you used when you had one or two branches, even if you transitioned from Leased Line to Frame Relay WAN links. So the obvious benefit of using Frame Relay instead of Leased Line is that you don't have to keep upgrading the router at the HQ whenever a new branch is opened. Another direct money benefit is that instead of three separate leased line service bills, you consolidate the three bills into a single Frame Relay service bill which should be relatively cheaper than separate leased lines.

Does this answer your questions or do you have any other questions?

Manglian

Tech-airman

Thanks for your detailed answer which involved hard work.

Leased Line

[Serial 0]
z
[Serial 0](Branch1)
(HQ)[Serial 1]
z
[Serial 0](Branch2)
[Serial 2]
z
[Serial 0](Branch3)

Frame Relay

+
z
[Serial 0](Branch1)
/
(HQ)[Serial 0]===============
z
[Serial 0](Branch2)
\
+
z
[Serial 0](Branch3)

This time it really makes sence to me. Ok at last, what about saving in terms of CSU/DSU ?. Per leased line how many CSU/DSU we use ?. To my understading answer is two.
Or if we use FR how many CSU/DSU we dont have to use ? under three branch - one HQ scenario

wbosher

Using leased lines, you would need a CSU/DSU on each end of each line (each physical interface). With FR you need one at each branch and one at HQ as you are only using one physical interface per router.

So you would only need 4 CSU/DSU's using FR, but you would need 6 if you had leased lines. Hope this makes sense.

kryolla

OP you need to understand point to point wan topology and multiaccess wan topology better then worry about csu/dsu etc.