Cisco Frame-relay Switch Configuration

This topic is to discuss the following lesson:

What happens to a dlci when it goes through a router? is this layer 2 from ISP to your remote site?

1 Like

Hello Justin,

As I was writing you answer I found that I need some deeper understanding of this topic myself. What I can say is that Rene has an excellent analogy for this.

This lesson may help you understand the topic better. I would dive into this lesson myself but I have just finished studying and need a break. Please let me know if I can help you on these forums!

I hope this helps,

Thanks for the link. I have read this article multiple times =)

1 Like

Hello Justin

Frame relay is a layer two technology and as a result, DLCIs do not traverse a router. In this example, the Frame Relay Switch is actually physically a router, a layer three device, but its functionality in this case is a layer two device. This can be seen clearly by the encapsulation of type “frame relay” that is specified in the config. Specifically it is switching between one DLCI to another, something that the ISP would do for the customer.

The only layer three devices here are the two customer devices. Now the DLCIs and their scope actually ends at the routers themselves, in a similar way to how MAC addresses have a scope within the specific network segment in which they reside.

So to specifically answer your question, this is a layer two connection between the customer and the ISP frame relay switch. Indeed it is a layer two connection all the way to the remote site as well. There is no layer three functionality that is intervening between R1 and R2. The switchover from DLCI 102 to 201 is an exclusively layer two operation.

I hope this has been helpful!


Thank you! It seemed that way its just so weird to me because the distance between two company sites can be very far. Thanks again for the help!

1 Like

Can I conclude that the frame relay technology has been deprecated because of its small bandwidth and non broadcast nature?


Hello Rocky

The reason why Frame Relay is declining in use is primarily because of the competition it’s facing from cheaper and faster technologies such as Cable, DSL, Metro Ethernet and MPLS, all of which, with the appropriate configuration, can outperform Frame Relay in most cases. Also, these technologies are more compatible with the LAN technologies used while Frame-Relay is completely different and requires the appropriate termination equipment (CSU/DSU etc) at the customer premises.

Limited bandwidth is not so much an issue as cost per bandwidth unit is. Getting a fast Frame Relay connection that is comparable to the aforementioned technologies can become prohibitively expensive. The non-broadcast nature is an issue, but can be dealt with relatively easily.

I hope this has been helpful!



Hello Rene,

Thanks for the great explanation.
The frame relay configuration posted in the tutorial is using local significance addressing.
Can you please provide example of frame relay configuration using global significance addressing ?


Hello Sachin

The mechanisms used for either globally or locally significant addressing are the same. The only difference is that with globally significant addressing, you are limited to using a particular set of DLCIs rather than choosing your own. Globally significant addressing is limited to using only 992 unique DLCIs since DLCIs are defined by 10 bits (2^10=1024) and 32 values are reserved DLCI addresses. This limitation has made globally significant addressing in frame relay an approach that is rarely used.

Nevertheless, you can find some information about globally significant addressing and how it compares to locally significant addressing at this Cisco documentation.

I hope this has been helpful!