How can QoS be handled for large IP networks and what are some of the implementation challenges?

How can QoS be handled for large IP networks and what are some of the implementation challenges?

I keep hearing that VoIP is not a proven solution for large networks mainly due to lack of QoS in IP. Some of the vendors suggest throwing bandwidth to overcome the QoS issues. How can QoS be handled for large IP networks and what are some of the implementation challenges?

    Requires Free Membership to View

    Login

    SearchUnifiedCommunications.com members gain immediate and unlimited access breaking industry news, expert advice on UC, technical guides, and more -- all at no cost. Join me on SearchUnifiedCommunications.com today!

    Kate Gerwig, Editorial Director

    By submitting your registration information to SearchUnifiedCommunications.com you agree to receive email communications from TechTarget and TechTarget partners. We encourage you to read our Privacy Policy which contains important disclosures about how we collect and use your registration and other information. If you reside outside of the United States, by submitting this registration information you consent to having your personal data transferred to and processed in the United States. Your use of SearchUnifiedCommunications.com is governed by our Terms of Use. You may contact us at webmaster@TechTarget.com.

IP has standard QoS mechanisms such as Differentiated Services (DiffServ) which can be enabled throughout an IP network. Proper implementation of QoS requires planning and the amount of planning is directly proportional to the size and complexity of the network. One of the more significant implementation challenges is designing to handle failures within a network where voice traffic will be routed along alternate paths. If the implementation is not well thought out, it is possible that voice traffic routed along alternate paths could cause the available bandwidth to be exceeded. The VoIP equipment that is chosen for deployment in a large network should have the ability to limit the number of calls set up between locations along the various pathways. With those limitations set, proper bandwidth calculations can be made to handle the various failure mode scenarios.

This was first published in July 2004