Printer Friendly

How frame relay stacks up against packet switching.

Despite some vendor and user reservations, frame relay (FR) is emerging as the leading candidate to interconnect dispersed LANs (local area networks). Its benefits are widely promoted by the major carriers and value-added network (VAN) service companies, as well as equipment suppliers.

AT&T and MCI are fueling the momentum by readying extensive FR services for introduction during the first half of 1992. They will join rivals Sprint and WilTel, who already offer FR services, along with VAN service providers BT North America, CompuServe and Infonet Services Corp.

Meanwhile, over 50 suppliers of multiplexers, bridges and routers have added support for FR operation or announced plans to do so. Most belong to the Frame Relay Forum, whose goal is the promotion, acceptance and implementation of FR, based on ANSI and CCITT standards.

Users have indirectly begun to promote growth of FR by demanding a cost-efficient way to handle the irregular bursts of high-speed data that characterize LAN-to-LAN traffic. With FR, applications can use the full speed of the access line for a short burst, then relinquish it for other applications to use. This dynamic sharing of bandwidth can provide savings of close to 50%, compared with traditional T1 muxing, according to carrier estimates.

FR vs X.25

FR is souped-up successor to X.25, providing higher-speed service with much less delay, while retaining the port-and bandwidth-sharing benefits.

Like X.25, FR provides multiple virtual circuits from a single access port, giving users a logical end-to-end link between all points connected to the service. Whereas X.25 sends data in packets, FR transmits (or relays) data in "frames," which can be as short or long as the user wants.

FR outperforms X.25 by reducing the switching delay through each network node by as much as 90% and by minimizing retransmission to correct errors. FR assumes the quality of today's fiber optic and digital links virtually eliminates the need for error detection and correction.

FR operates at layers 1 and 2 of the seven-layer OSI model. In contrast, X.25 also defines layer 3, which carries out flow control, validation and error correction. FR relies on layer 3 and higher OSI protocols built into customer premises equipment for retransmission of lost frames or incorrect data and to recover from congestion.

Since FR only involves the lower two OSI layers, network protocols requiring three or more layers, such as TCP/IP, SNA, Ethernet, token ring and FDDI, can run transparently over a FR network. This ability to carry SNA traffic in conjunction with a variety of LAN traffic adds to FR's appeal for interconnecting LANs and tying into WANs (wide area networks).

Concerns remain

Even so, questions remain. What if the quality of the links is suspect? A faulty network infrastructure will severely impair performance because of the need for retransmission.

Also, because FR depends on the end system for frame validation and error control, a lost frame will not be detected until the incomplete data reaches the destination. Protocols providing the functions of layers 3 and 4 of the OSI model, in contrast, detect and correct the loss as soon as it occurs.

Further, when the end system detects congestion, it must send a frame with the message back to the source. Not only does this add to congestion, but the frame risks being lost in transit.

Users burned by the changes in the early days of X.25 are wary of making the same mistake with FR. X.25 systems built to specs in 1976 were unable to operate with those built four years later.

FR standards are maturing faster than X.25 did, but there have already been a number of revisions, and differences remain in national and international specs. Users still need to be cynical of equipment interoperability claims from different manufacturers. Also, the carriers have not interconnected their FR backbones, so users are effectively restrained to using a single carrier.

Faced with these concerns and today's economy, users appear reluctant to invest in FR. A recent study by Vertical Systems Group of Dedham, Mass., turned up only 30 or so commercial FR users, with a similar number in the deployment stage.

To encourage more organizations to embrace FR, some vendors are using free trials as a lure. WilTel has joined Wellfleet Communications, General DataComm and Cisco Systems to provide the long-distance services and equipment needed for a free-30-day trial. Other carriers are promoting one-stop shopping by testing and certifying various manufacturers' equipment and interoperability with their networks.

If questions of compatibility among services and hardware can be resolved, along with the issues of congestion management, FR should flourish--at least until the mid-90s when even higher-speed services will be available.
COPYRIGHT 1992 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Network Management
Author:Edwards, Morris
Publication:Communications News
Article Type:Column
Date:Apr 1, 1992
Previous Article:20 ways to assure you're not framed by frame relay.
Next Article:Another hat to wear: security guard in an ISDN world.

Related Articles
Fast packet networking strides ahead.
Two users discuss pros and cons of frame relay.
Vendors reply to frame relay 20 questions.
A formula for the future.
ComNet, like the Olympics: stronger, faster.
Video follows voice over frame relay.
Video follows voice over frame relay
It's elementary, my dear Watson.
Optimize with SVCs.
Is the clock ticking for frame relay?

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters