Printer Friendly

Is best effort good enough? Good enough for what? (Guest Column).

Is best-effort IP service good enough? I suppose the answer depends on what you are trying to accomplish. This question always shapes up as a debate between the "big fat pipe" guys vs. the service-level guys. Instead of debating whether best-effort IP service is good enough, however, we should be asking, "Good enough for what?"

Is best-effort service good enough to build a robust global data network?

Absolutely. We have an existence proof called the Internet. The inherent best-effort nature of IP is one of its most powerful strengths. Simple and scalable, best-effort delivery is a cornerstone principle that has enabled the expansion of the Internet to the far reaches of the world. Arguably one of the most significant developments of the 20th century, the Internet has proven that data networks can be scaled to the same global proportions as the public-switched telephone network.

Is best-effort service good enough to build a profitable business?

No. Internet service providers, or carriers, are not necessarily able to define and execute a durable and profitable business model based on best-effort Internet service. This is not a criticism, but finding an ISP that is not struggling with its business model is difficult.

Consumers of all types attribute greater value to products and services that offer guarantees, rather than to those that do not. By direct extension, this means that best-effort service will always be viewed as less valuable when compared to technologies that offer service assurances.

Service providers must evolve their IP service offerings beyond best effort if they hope to capture any share of the lucrative business markets. They must establish a clear and unambiguous value proposition that clearly delineates between best-effort service and guaranteed service levels if they are ever to stand a chance of achieving a reasonable return on invested capital associated with their IP infrastructure.

Ironically, most service providers have plenty of network capacity and the required technology, such as MPLS and DiffServ, at their disposal to secure a meaningful class-of-service hierarchy with service level guarantees. Forging ahead will require sustaining repeatable service-level parameters, such as committed bandwidth, service availability, loss, latency and jitter. These service-level parameters need to be verifiable by the subscriber and should result in establishing a compelling value proposition.

Is best-effort service good enough to support time-sensitive traffic?

Here, again, the answer is no. Best-effort delivery is not adequate for supporting time-sensitive services that require consistent performance with respect to loss, delay and jitter. In most cases, to achieve predictable performance levels, a connection-oriented technique like MPLS is required to reserve and manage network resources.

Services such as virtual leased lines, toll-grade voice, streaming media and network-attached storage--all of which are targeted to run over IP infrastructure--require class-of-service commitments that go beyond the capabilities of best-effort services. These services represent potential new sources of revenue that ISPs and carriers cannot afford to miss.

Is best-effort service good enough to support non-time-sensitive traffic classes?

Yes. Services such as e-mail, Web surfing and messaging can all effectively use best-effort delivery mechanisms to accomplish their communications objectives.

While this list of questions is by no means exhaustive or fully encompassing, it is clear that best-effort service is good and not good enough. It absolutely depends on what it is used for.

I can't help but ask one more question, however, one that is very important given the massive investment made in IP infrastructure over the last five years. What technology stands ready to enhance the value of IP and extend its use in more than best-efforts applications?

MPLS and DiffServ are key technologies that can--and will--extend and enhance the value of IP networks with respect to supporting all forms of service traffic across a variable-length packet backbone. Inspired and shaped from the same people that made IP a household phrase, MPLS and DiffServ stand ready to bring IP to its next level of greatness.

Welch is senior vice president of marketing and business development at Tenor Networks, Acton, MA. Send comments to
COPYRIGHT 2002 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Comment:Is best effort good enough? Good enough for what? (Guest Column).(Column)
Author:Welch, Sean
Publication:Communications News
Article Type:Column
Geographic Code:1USA
Date:May 1, 2002
Previous Article:Fiber optics make the grade. (Trends).
Next Article:Up to the test: Chris Odell guides fluke networks in delivering network analysis tools. (Cover Story).

Related Articles
Media relations: getting it right.
Guest columns: readers respond.
Reading stock tables is fundamental.
Perfect vs. good enough. (Editor's Corner).
Well, good enough, it's perfect. (Editor's Corner).
When it Raines. (Here Below).
Taming the media: how one CPA turns media savvy into business opportunities.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters