ITPRC NEWS - May, 2002 - http://www.itprc.com/

Voice over IP We hardly knew ye'
By Irwin Lazar

This week's Network+Interop conference was notable for what wasn't covered in great detail as opposed to what was. Major focus areas of vendors included products and services for content delivery and web-application optimization, wireless LAN services, and network security. What wasn't prevalent at this show was Voice over IP. What happened?

A year ago the future of networking was the vision of a converged network capable of supporting voice, video, and data via a single unified infrastructure. While that vision remains unchanged for most service providers and vendors, there is a small but growing number of individuals questioning this approach, and even more importantly, a marked lack of decline in interest in this topic among both enterprises and service providers.

Why?
There are three factors driving the decline in momentum for VoIP cost, maturity, and management.

Cost - VoIP was touted years ago as a way of saving money. By leveraging the existing data network infrastructure organizations could eliminate much of their phone bill. In addition, convergence on a single unified infrastructure offered cost savings by eliminating redundancy and the cost of operating separate networks for voice and data. The reality that many organizations have discovered is that many current data networks aren't ready for voice. They lack the resiliency and QoS features that are required to deliver voice at an equivalent level of existing services.

Maturity -  Organizations are discovering that VoIP solutions are still maturing. Call manager applications are frequently upgraded to add new features, yet the feature list of many VoIP solutions is still not as broad as that which one can find in a traditional PBX or key system. Embracing VoIP in the current environment may mean a loss of functionality, and will most definitely require many upgrades as feature sets mature.

Management -  Managing voice is unlike managing most applications. VoIP requires end-to-end application performance management capabilities that many organizations lack. Call quality is dependent on a number of factors, including congestion, latency, jitter, and server load. Supporting voice services requires the tools, systems, and methodologies to troubleshoot the entire conversation path. For network managers, the days of saying "the T1 is up" are over the minute that they decide to deploy VoIP.

The Future
Does this mean that VoIP is dead? Hardly. Organizations still view VoIP as their future method of delivering voice. SIP and its accompanying architecture may enable a new class of services that will drive demand for VoIP. For new installs where networks can be deployed with resiliency and performance features to guarantee voice quality, VoIP still makes sense. For organizations with a large legacy network that requires significant investment to support voice, the justification for VoIP may not exist.

As we've seen numerous times in this industry, the hype curve often far exceeds reality. Migration to VoIP will be a long and slow process. Many organizations may decide it isn't necessary; most will still implement it in some fashion. However, it just may take longer than we thought.

For More Info:
The ITPRC "Voice and Data" page contains dozens of links to information on voice and data convergence. Visit it at www.itprc.com/voice.htm?newsletter 

..............................
Irwin Lazar is a Senior Consultant for Burton Group where he focuses on strategic planning and network architecture for Fortune 500 enterprises as well as large service providers. He is the conference director for MPLScon and runs The MPLS Resource Center and The Information Technology Professional's Resource Center. 

Please send any comments about this article to ilazar@burtongroup.com ============================================================