Introduction

Frame Relay is a packet switching technology, that relies on low error rates, digital transmission links and high performance processors. It was intended to be an intermidiate solution for the rapid increase in the demand of high bandwidth communication (e.g. LANs) and was originally conceived of as a protocol for use over ISDN interfaces.
Initial proposals to this effect were submitted in 1984 to the Consultative Committee for International Telegraph and Telephone (CCITT). Work on frame relay was also undertaken in the American National Standards Institute (ANSI)-accredited T1S1 standards committee in the U.S.

Frame Relay technology was designed with the following features in mind:
 

1. Low latency and higher throughput - To achieve this a simple link layer protocol is used.

Note: Frame Relay exists only at layer 2. This means that it has significantly less processing to do at each node. therefore improving throughput by orders of magnitude (vs. X.25).
2. Bandwidth on demand - It is highly desirable that the user be assigned the bandwidth based on the actual demands. The bandwidth can be allocated to users at call setup. However, to make it more effective, a user can renegotiate the requested bandwidth whenever it has to transfer traffic bursts with certain peak rates throughout the call.

Note: This differs from X.25, which has a fixed bandwidth available (it uses or waste portions of its bandwidth as the load dictates).
3. Dynamic sharing of bandwidth - Increased sharing of resources would yield a better utilization of the network bandwidth. The bursty nature of data traffic could be exploited by allowing some users to consume the bandwidth during other users idle periods.
4. Backbone network - A backbone network is needed to maximize user connectivity, to accommodate a variety of end system technologies, and to be less vendor dependent.

Applications in Use

Applications which are particularly suited to use the Frame Relay protocol are applications that:

1. Require the consolidated transport of several protocols.
2. LAN to LAN interconnections and other applications that generate bursty traffic.
3. Support Large Host computers by providing a cost effective multiplexed communications interface, e.g. SNA transport.

History of Implemetation

Public Frame Relay networks initially took off in North America.
Packet switching services in North America, although available, were not generally popular with the users since they were seen as providing slow and low throughput networks. As a result many companies installed their own private networks, based upon time division multiplexing principles, over inexpensive and high bandwidth leased lines. They justified the existence of these private networks purely on cost savings made by passing their voice traffic over their private network rather than a public network.

The network operators in North America fought against this. They offered virtual private network capabilities for voice traffic. Thus providing a secure, but comparatively low cost, voice network within the domain of the public network operator. This left the private companies with the requirement to pass data efficiently across a private network.
The newer style of data applications is 'bursty' in nature and requires access to the full network bandwidth on demand. Time division multiplexers are unable to offer this capability. They provide the user with a permanent portion of the bandwidth, rather than temporarily providing the full bandwidth.

The public network operators saw an opportunity to take back the data networking they had lost to the private networks.
Frame relay had been tariffed in North America extremely aggressively and was targeted at competing with the tariffs for leased lines. The first public frame relay service offering appeared in North America in 1992. Companies such as AT&T, US Sprint, BT North America, Wiltel and Compuserve installed frame relay nodes in major cities throughout the United State and permitted access to these nodes (at the POP - point of presence) by a local access line purchased from the local telecommunications company. The customer had to pay for the subscription to the frame relay service and for the access line from the local telecommunications company (normally at a speed between 56 kbps and 1.544 Mbps). For customers within the major cities, where the public frame relay operator's nodes were situated, this was not a major problem since the cost of the access line was fairly small.

In Europe the situation was slightly different. European public network operators had not provided high-speed leased lines at a low cost. As a result the demand for time division multiplexers had been lower. Furthermore, the public network operators had not provided good quality, high-speed public packet switching services (apart from France). The consequence of this was that the market for private data networks (based upon X.25 packet switching at lower line speeds) had been enormous. Companies of all sizes used private packet switching networks. Since packet switching in general, and X.25 in particular, were seen as extremely cost effective private networking mechanisms, the potential market for public frame relay services in Europe was much smaller than in North America.

In Europe the only service providers to offer frame relay services during 1992 were BT, with extensions to their Global Network Services (GNS) managed Packet Switched data network service, and the Finnish PTO, which installed a limited frame relay network in Finland.

Frame Relay vs  X.25

Frame relay is a streamlined packet transfer method of X25.
It is a switching and statistical multiplexing technology without the error control of the X.25, therefore being much faster.
While X.25 is only implemented at speeds below 64 Kbps, frame relay is implemented up to T1 (24 times faster) and some carriers may implement it at T3 rates (672 times faster) - therefore it's sometime called fast packet.
X25 was created with the intention to operate up to the 3rd level of the OSI model. Frame Relay only operates at the first two layers of the model.
One can say that Frame Relay is basically "dumb" and relies upon customer equipment or DTE gear to monitor and govern flows and do error correction. Due to the fact that Frame Relay typically operates over WAN facilities that offer more reliable connection services, fewer of the robust capabilities of the X.25 (such as windowing and retransmitting of lost data) were worked into the Frame Relay protocol. This means that frame relay has significantly less processing to do at each node, which improves throughput by order of magnitude.
This enables Frame Relay to offer higher performance and greater transmission efficiency than X.25 and makes Frame Relay suitable for current WAN applications, such as LAN interconnection

More Comparison points of X25 and Frame Relay:

More About X.25:

The most important thing to remember with X25 is that it was a predecessor Frame Relay. It is a packet switching technology, that is, you send data packets through an access line into the network cloud just like you send soap bubbles through a straw. Each packet is labeled with an identifier (all your packets during this session will have the same identifier). The network sends all the packets with the same identifier via a permanent virtual circuit (PVC) or switched virtual circuit (SVC). The packets are delivered through the other access line to the X.25 destination, and, of course, packets can go both ways (full duplex).
Every node of the network performs an extensive end-to-end error control and, if necessary, transmissions are retried several times. Making X.25 a tedious and slow but very thorough.

X.25 was designed to provide error-free delivery using high error-rate links. X.25 prepares and sends packets, the packets contain several fields used for error and flow control. X.25 has a fixed bandwidth available. It uses or wastes portions of its bandwidth as the load dictates.

Today's networks don't need all these smarts. They are faster, bigger and dumber, but maintaining the essential bandwidth-on-demand characteristics of the X.25.