Why oversubscribe, anyway?

Continuing our discussion of oversubscribing frame relay ports, the obvious question is whether anybody would want to oversubscribe. Isn't there a possibility that the port might not be able to handle all of the data that needs to flow across a PVC?

The answer is "yes." There is a possibility that enough bandwidth might not be available at the port, but the odds of that happening are sufficiently slim. And with the savings being sufficiently great, it might be a good idea anyway.

First, a couple of clarifications. The frame relay committed information rate (CIR) applies only to permanent virtual circuits (PVC), and has nothing to do with ports. This makes sense because ports are dedicated to a single customer. The port bandwidth, then, is dedicated to particular customer. Therefore, there is no need for a CIR on a port; by definition, the port speed is a "committed" speed.

On the other hand, CIR was invented as a safeguard against a carrier's oversubscribing the network bandwidth across its backbone excessively. The CIR is the carrier's guarantee that the network bandwidth will be available for that PVC if and when the customer needs it.

So, why would any sane person want to take the risk of oversubscription and potential port bottlenecks? Consider this situation. A company needs the New York office to be able to communicate with six other sites (Atlanta, Chicago, San Francisco, Denver, Omaha and Los Angeles) at 6K bit/sec. Whenever communicating with any of the sites, 64K bit/sec is needed, so a 64K bit/sec PVC is assigned for each. But the odds are very slim that 64K bit/sec will be needed to all six sites at the exact same moment.

There are two solutions to the above situation. On one hand, the network could be designed with a conservative belt-and-suspenders approach of six PVCs with a 64K bit/sec CIR for each and a 384K bit/sec port. This would also require a full T-1 access line, since fractional T-1 is not usually available for local loops. On the other hand, the same six PVCs with the same 64K bit/sec CIR possibly could be supported with a single 64K bit/sec port and access line.

Exactly how much would this save? Is it enough to be worth the risk? Stay tuned for our next newsletter for those calculations.

---

Steven Taylor, consultant and broadband packet evangelist, and Joanie Wexler, an independent networking technology editor and writer, team up to bring you this analysis and commentary. Taylor specializes in education and market analysis, and Wexler adds incisive reporting and research. For more detailed information on most of the topics discussed in this newsletter, connect to www.webtorials.com, the first Web site dedicated exclusively to market studies and technology tutorials in the Broadband Packet areas of Frame Relay, ATM, and IP. Feedback and additional topic ideas are welcome. Please contact taylor@webtorials.com or joanie_wexler@mindspring.com>

Frame relay at the crossroads
Network World, 05/31/99

Suppressing silence saves money
Network World, 02/01/99

Trillium frame relay white paper
Trillium Digital Systems, Inc.

Archive of Network World on Frame Relay newsletters