cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
8392
Views
20
Helpful
7
Replies

Questions on Leased Line

SecureBlob
Level 1
Level 1

When a company gets a leased line connection between 2 offices in 2 different cities would there be a dedicated line all the way from the office in first city to the office in 2nd City? Is the below representation accurate? "-->" represents a Leased Line.

 

Office in City A-->ISP office in City A-->ISP Office in City B-->Office in City B

 

If that is true, whenever a new customer requests for a leased line, ISP would need to install the leased line from one location to the other location. Am i right?

 

Another question is about the Layer 2 Encapsulation protocol that can be used on the leased line. It could be PPP, HDLC or SLIP. Am i right?


If that is true, why is the link below says "...if you have a T1 leased line to the Internet or a private network between locations, you use one of these three WAN Protocols: HDLC, PPP, or Frame-relay."

 

https://www.petri.com/csc_3_wan_protocols_you_should_know

 

How come the author is talking about Frame-Relay in the context of Leased Line. Does it mean that Frame-Relay can work on a Leased line? If the answer is yes, does it mean that all other packet switched Layer 2 encapsulation protocols can work over a leased line?

1 Accepted Solution

Accepted Solutions

Correct, a circuit is what's dedicated between the end points.  However, the local line, from the customer's premise to the telco (provided by the telco) is/was often dedicated to the leased line.  Once the circuit gets into the telco's network, it runs across lines being used for other circuits.

 

Yes, today, trunk lines might be copper or fiber, although today telcos are more likely to use fiber for those links.  Also the "circuit" is probably digitally shared within the telco's network, where once upon a time, circuit bandwidth was dedicated to the leased line end-to-end.

 

BTW, don't confuse a WAN provider with an ISP although they may both be the same company.  The former gets you a connection to somewhere, the latter provides a connection to the Internet.

 

As to L2 encapsulation, that depends on the nature of the circuit being used.  When WAN circuits were basically analog phone circuits, that we used for digital data, then those encapsulations were commonly used.  Today, in modern networks you're may be more likely to encounter Ethernet hand offs, or other high speed digital hand offs.

 

 

View solution in original post

7 Replies 7

Hi @SecureBlob

Your questions make sense however I need to call your attention to the fact that leased line has broad definition.

 The line is the physical media and what exactly it is depends on what  ISP offer.

If you or ISP install a router or another device (Frad) on your facility and allows you to build up a frame relay over it, it is ok. 

 The encapsulation to be used depends on the device on both ends.

 Frame relay is just another layer 2 wan protocol. The difference between FR and those you mentioned is the ability of FR to create virtual circuits. But this is a different thing. 

 In terms of physical connection, it all depends on the device that will receive the leased line.

 

Joseph W. Doherty
Hall of Fame
Hall of Fame
A true (p2p) leased line is often really a dedicated circuit, end-to-end. I.e. it's like making a permanent phone call. As to physical wiring, each end might get a dedicated physical link from the local exchange to the customer's premise. (Between exchanges, the circuit would usually ride across trunks.)

As to frame-relay, leased lines are generally used from the customer's premise to the frame-relay provider's frame-relay network.

Joseph

I have a follow up question. You said "A true (p2p) leased line is often really a dedicated circuit, end-to-end". Does it mean that there can be a leased line but it might not have a dedicated physical line? In that case, how is it working like a leased line?

 

After i posted my original question, I saw the link below.

https://www.lifewire.com/definition-of-leased-line-817871

 

It says, " A leased line is not a dedicated cable; it is a reserved circuit between two points. ". But many links say that a leased line is a dedicated connection. What is true.

 

You said "Between exchanges, the circuit would usually ride across trunks." So is the below representation correct?

 

Office in City A-->ISP office in City A=====Trunk lines (shared by many customers)========ISP Office in City B-->Office in City B

 

I assume that these trunk lines can be over copper or fiber. Am I right?

 

Flavio/Joseph/Anyone else, 

If the connection is a true p2p leased line connection between  Office in City A and Office in City B, can you confirm that the the Layer 2 Encapsulation protocol should be PPP, HDLC or SLIP. Am i right?

 

 

Only time Frame relay comes into picture is when the connection is like below.

Office in City A ===> Frame Relay Clould ===> Office in City B.

The connection to Frame Relay can be dedicated line. 

 

My last question is, what are the other options possible other than a dedicated line to ISP? Is it possible for multiple customers to share connection to Frame Relay clould? 

 

Hi,

I think your question mix different concept,  I try to address every question:

 


I have a follow up question. You said "A true (p2p) leased line is often really a dedicated circuit, end-to-end". Does it mean that there can be a leased line but it might not have a dedicated physical line? In that case, how is it working like a leased line?

p2p has to do with topology,  the fact the a point-to-point connection is not shared has to do with the kind of transport network. Until 5 to 10 years ago you can have primarily three kind of connections:

 

ATM/Frame Relay network: in this case you have a dedicated access link connecting some kind of CPE (usually router but before routers frad were used) to a FR or ATM switch. After this your frames/cells were delivered using a shred network this means that links connecting ATM/FR device are not use just for you data. Bandwidth is shared.

 

Clear Channel or PDH/SDH links: in this case you have a real dedicated circuit, this mean the ISP give you a reserved path over its network; the amount of bandwidth allocated to you can't be used by anyother. If you don't use it it remains empty. That's way this kind of link were much more expensive. In this case you can choose the layer 2 protocol  you prefer: PPP, HDLC,...

 

MPLS networks: in the last 10 years MPLS network became much more important. MPLS networks are shared networks but they can also emulate a point-to-point link that can be very much similar to a clear channel (in some case they are not "transparent" and some kind of protocols doesn't flow over the net)

 


@SecureBlob wrote:

Joseph

I have a follow up question. You said "A true (p2p) leased line is often really a dedicated circuit, end-to-end". Does it mean that there can be a leased line but it might not have a dedicated physical line? In that case, how is it working like a leased line?

 

After i posted my original question, I saw the link below.

https://www.lifewire.com/definition-of-leased-line-817871

 

It says, " A leased line is not a dedicated cable; it is a reserved circuit between two points. ". But many links say that a leased line is a dedicated connection. What is true.

Correct, ISP never connect to sites directly, so you will never have a wire going from one to the other. This is due to two main reasons:

  • electric (and also optical) signals must be regenerated because as far as they travel over e physical medium, they fades
  • also in case of very narrow sites, ISP connect you to some kind of active device in their POP otherwise they can't monitor the link. 

 

"If the connection is a true p2p leased line connection between  Office in City A and Office in City B, can you confirm that the the Layer 2 Encapsulation protocol should be PPP, HDLC or SLIP. Am i right?"

p2p is topological term while you are asking something that has to to with ISO-OSI layer 2.  If the link is a clear channel you can use whatever Layer 2 protocol you prefer. If it is a p2p connection build over someother kink of transport networks the answer depends on the kind of network, device and configuration done by the ISP.  Usually Ethernet is used. Inany case you must ask your ISP.

 

 

"Only time Frame relay comes into picture is when the connection is like below.

Office in City A ===> Frame Relay Clould ===> Office in City B.

Correct

 

The connection to Frame Relay can be dedicated line.

Yes, but you have a dedicated link just from your site to the ISP then it's a shared network

 

My last question is, what are the other options possible other than a dedicated line to ISP?

Yes, has I wrote the most common kind of connection is true an MPLS network or, in some case, over an Ethernet network.

 

Is it possible for multiple customers to share connection to Frame Relay clould? 

Yes, that's what FR netwrok are made for. The real question is if they can exchange data or not and this it's depends on the way the ISP configure the network. Usually each customer has its own private network (VPN) but ISP can configure MPLS networks to share traffic.

 

Bye,

enrico

 

PS: rate if useful

 

 

Correct, a circuit is what's dedicated between the end points.  However, the local line, from the customer's premise to the telco (provided by the telco) is/was often dedicated to the leased line.  Once the circuit gets into the telco's network, it runs across lines being used for other circuits.

 

Yes, today, trunk lines might be copper or fiber, although today telcos are more likely to use fiber for those links.  Also the "circuit" is probably digitally shared within the telco's network, where once upon a time, circuit bandwidth was dedicated to the leased line end-to-end.

 

BTW, don't confuse a WAN provider with an ISP although they may both be the same company.  The former gets you a connection to somewhere, the latter provides a connection to the Internet.

 

As to L2 encapsulation, that depends on the nature of the circuit being used.  When WAN circuits were basically analog phone circuits, that we used for digital data, then those encapsulations were commonly used.  Today, in modern networks you're may be more likely to encounter Ethernet hand offs, or other high speed digital hand offs.

 

 

SecureBlob
Level 1
Level 1

Here is a beautiful video that talks about how leased lines work.

https://www.youtube.com/watch?v=paoL74cZivI

 

I am attaching a picture of that too. The link between the MUXs are shared. Leased Line.png

 

"The link between the MUXs are shared."

 

Yes but I think that the real point is not if the link are shared but how they are shared. In PDH/SDH network you multiplex many lower bandwidth links, over a higher bandwidth link; I mean that an ADM (the MUX in SDH nomenclature) multiplexes, for example, many 2Mbps links (so called E1 links) over a single 34Mbps link (so called E3 link). In this scenario the network shares the link but the link itself is logically subdivided in many "channel" (note that it's not the correct way to cal them) and the bandwidth associated with each one is not shared. If you don't send data, that amount of bandwidth remains unused and that's way this kind of link are much more expensive then any kind of shared network. Infact the Telco can't make any kind of sharing. If you have 16 E1 links multiplexed over an E3 link, you need whole E3 link (note that 16xE1 is 32Mbps but you have an extra 2Mbps for signalling and control). If you need 2Mbps between to sitethat are 1000Km far, the the Telco have to reserve 2Mbps over the whole backbone  connecting that sites. In shared network, a Telco can assume that is highly unlike that all the E1 link are simultaneously forwarding data at full speed, and make some kind of overbooking (I'll explain below) 

 

In other networks (MPLS for example) the bandwidth at some point, is shared ; for example in a MPLS network you are connected to a so called PE router (the edge router of a MPLS network). Usually you have a dedicated link (meaning tha you have your reserved amount of bandwidth) form branch to PE. PE's are connected each other through P routers (the backbone router of the MPLS network) using links (see below).

 

03fig03

 

Bandwidth between PE to P routers and connecting P to P routers, is shared meaning that there is not reservation.So if PE1 is connected to P3 with a 10Gbps link, you could connect, for exaple, 11 1Gbps links to PE1 bacuse its unlike that all of them are simultaneously forwarding data at full speed and, in any case they are, it will be for a short amount of time so PE1 could buffer. In theory, in case of congestion, your packets could be discarded due to congestion. This can't happen in an old SDH network.Obviously Telco job is monitor links to avoid this.

 

So sharing a physical medium doesn't imply sharing the bandwidth, the point is if and how the bandwidth is reserved.

 

By the way, so far we haven't highlighted a point that can be very important: jitter and delay.  Going back to the old SDH link: in this scenario delay is very low because SDH devices don't route or switch packets in the meaning we use these terms  speaking of routers and switches. Moreover they can't have congestion so jitter tend to be 0.  Over a routed or switched network, each device have to check some kind of address inside the frames/cells/packets to forward them. This job requires time and, because they can have congestion, delay can change over time and that's means jitter.

 

A final consideration: the above explanation is just to give you an idea, the real  way networks work is quite complex and, to be explained in detail, require books, not post ;-)

 

Bye,

enrico

 

 

 

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: