Frame Relay is an industry-standard switched data link layer protocol operating at the physical and data link layers of the OSI model. It can handle multiple virtual circuits (VCs) between Frame Relay-capable devices.

The fields of the Frame Relay frame are as follows:

• Flag Used to identify the beginning and end of a frame.

• Data-link connection identifier (DLCI) Identifies the path through the network to the destination.

• Command/response (C/R) Not generally used.

• Extended address (EA) Identifies whether the header octet is followed by another header octet. A value of 0 means that another octet follows. 1 identifies the last octet.

• Forward explicit congestion notification (FECN) Used to inform the connected devices of congestion in the network.

• Backward explicit congestion notification (BECN) Used to inform the connected devices of congestion in the network.

• Discard eligible (DE) Used to identify a packet that is eligible for discard during congestion.

• Data Can be used to carry any type of information.

• Cyclic redundancy check (CRC) Used to detect transmission errors and cover the header and data fields.

The 10-bit DLCI value is a logical number in the range of 16 to 107 used to identify the logical connection or permanent virtual circuit (PVC) that will be multiplexed into the physical circuit. The DLCI has significance only between your customer premises equipment (CPE) and your provider's Frame Relay switch. Because the DLCI is used to differentiate between different conversations on the same physical circuit, you can think of it as the heart of the Frame Relay header. Without it, your Frame Relay access device (FRAD) could not identify the different data streams passing through it.

Frame Relay provides you with a packet-switching data communications capability that is used across a data network interface to identify how the traffic will be formatted between your devices (such as routers, switches, multiplexers, and concentrators) and your service provider's network equipment (such as Frame Relay switching nodes). You need to know a couple terms used in Frame Relay. Your devices are often called data terminal equipment (DTE), and your service provider's network equipment is often called data circuit-terminating equipment (DCE).

As the interface between the DTE and DCE, Frame Relay must provide a technique that can statistically multiplex many logical data conversations over your single physical transmission link. If you are familiar with systems that use only time-division multiplexing (TDM) techniques to support multiple data streams, this technique might seem alien to you. Because Frame Relay was conceived to replace less-efficient protocols, its statistical multiplexing provides more flexible and efficient use of available bandwidth than a traditional TDM circuit. You should be aware that you can run Frame Relay on top of the channels provided by a TDM circuit, or you can run Frame Relay without any TDM techniques.

Frame Relay is a high-performance WAN protocol. It is standardized in the U.S. as an ANSI standard and internationally as an International Telecommunication Union Telecommunication Standardization Sector (ITU-T) standard. It operates at the physical and data link layers of the OSI reference model, much like Ethernet and Token Ring.

Frame Relay was created to develop the next-generation protocol to replace X.25 that would be carried across an ISDN interface, but it has been adapted to operate successfully over a wide variety of other network interfaces as well. Originally, WAN technology was developed to operate over low-quality transmission lines. Frame Relay can exploit the recent advances in WAN transmission technology. Because the earlier transmission lines were predominately analog transmission facilities, protocols such as X.25 were overengineered with extensive error checking and correction techniques to combat the quality of the communications across copper transmission lines. Although Frame Relay does not implement error checking, it does frame error checking and sends any error information to upper-layer protocols for any necessary actions, such as a TCP retransmission.

Today's links are much more reliable, often running across fiber media/digital transmission links. Because of this, Frame Relay can leave error detection and correction up to the higher protocol layers. Frame Relay does include a CRC algorithm for use in detecting corrupted bits so that the data can be discarded, but it does not include any protocol mechanisms for correcting bad data.

Frame Relay does not need to provide the explicit, per-VC flow control that X.25 implements. In its place, Frame Relay uses a simple congestion-notification mechanism that allows a network to inform a FRAD that the network resources are close to a congested state. This notification can also be used to alert the higher-layer protocols that flow control might be needed.

By using Frame Relay, you can reduce your network complexity and simplify your network architecture by supporting the three-tiered network model of core, distribution, and access layers. Frame Relay supports many different topologies for the placement of your network equipment, including full, partial, and hybrid meshing