If you're planning a VoIP strategy for a large organization, you may be wondering whether or not you should standardize on a codec. Standards, of course, are extremely important if you have to manage a large environment. For instance, if you have 1000 users and each of them have a different make/model of PC, with random versions of operating systems, it would be a nightmare to support. So standards are good, but they also necessarily restrict flexibility so the standard may be acceptable for all users but it's rarely optimal for any particular user.
So should you require everyone in your VoIP environment to use a particular codec? As usual, the answer to this question depends on your situation, but here are some of the pros and cons:
If all VoIP calls have the same maximum bitrate, then it's very easy to determine your bandwidth requirements. If you have a random mix of bitrates, it's almost impossible.
A standard makes configuration and your documentation easy.
A standard almost guarantees all devices can interoperate.
A standard means, in theory, that all users should experience similar quality. If users variously connect with different codecs which have different MOSs, they may lose the ability to distinguish between having a problem and using a low-quality codec.
A standard won't be appropriate in every circumstance. You may not want to use a high-bandwidth codec on your WAN, or a low-quality, low-bandwidth
A standard may prevent you from using a particular end-point if it doesn't support the standard.
When you make this decision, consider identifying regions of your network and choosing a standard codec for each region. This compromise adds a little flexibility, but maintains manageability. For instance, you might choose one codec for calls inside your LAN, and another codec for IP phone-to-IP phone calls that cross your WAN, and perhaps a third codec for PBX-to-PBX toll-bypass calls on your WAN.
Thomas Alexander Lancaster IV is a consultant and author with over ten years experience in the networking industry, focused on Internet infrastructure.
This was first published in April 2003