Desktop video conferencing brings the power of visual communications to users' PCs. There are a number of advantages
to this model of video conferencing including the low cost of the endpoint, no requirement for scheduling equipment or a room, the single-face per screen format, the ease of integrating with a unified communications environment and with data collaboration, and more. But of course there is no free lunch, and someone has to tackle the challenges that this deployment brings with it.
Integrated or standalone?
Desktop video conferencing can be deployed as a standalone system, as a system that integrates with existing room-based systems and/or telepresence systems, or as a service integrated with a larger unified communications (UC) service. This decision is largely driven by the circumstances of each enterprise. If you have no existing video conferencing and expect to be just desktop-based, look for one of the low-cost vendors or services that provide this capability. If room-based and telepresence systems are deployed and the business use requires integration, then look for a supplier who solves this integration task well. Conversely, if a UC environment exists or is planned, look into UC and video conferencing integration and make sure the resulting video service will have all the features you require for your specific business needs.
Video conferencing requires real-time compression and decompression of visual information, which is a CPU intensive task. Telepresence and room-based systems have traditionally addressed this with dedicated hardware. PC-based endpoints rely on the PC's processor to accomplish these tasks. Older laptops or desktops may not have the horsepower to keep up with both a high quality video conference and the other applications a user wants to run in parallel. Test your typical desktop deployment to ensure your PCs are up to the task.
Unlike a telephone call where we manage acoustics by putting the speaker right next to our ear and the microphone right in front of our mouth, video conferencing is dependent on the environment in which it is running. Video needs sufficient light on faces to show details, but not too much light like an outside window with the sun streaming in. The contrast of sunlight is too much for a video camera and participants look washed out. Likewise, if audio is using a speakerphone-like device or the embedded microphone and speakers in the PC, the noise of the environment will impact the conference. Are your typical offices acoustically and visually ready to provide a good video experience?
Echo cancellation or headsets
Echo cancellation is not needed when you put a telephone handset to your ear, but it is needed for a speakerphone-type environment. Headsets can be used to avoid echo, but users may not be pleased about always wearing them. Make sure that your desktop video conferencing solution has good echo cancellation technology, and that it works well with the PCs and/or appliances you choose to use as well as the acoustics of your typical office environment.
Most current desktop video conferencing solutions are server based, with some kind of portal acting as a directory for setting up a call. This same server verifies that the right software is loaded on the PC and that it can connect to the audio and video peripherals. This is definitely the way to go in a large endpoint deployment. The older method is to have stand-alone software that is installed on each PC, which means that software upgrades are painful. A server-based solution provides centralized management and reduced administrative overhead.
As soon as users realize they can have video on their PC, someone wants to travel with it. Often, travelling employees connect back to the enterprise through a VPN tunnel. The tunnel termination device may be taxed if a number of concurrent video conferencing calls are coming in from the Internet. Ensure you have sufficient Internet bandwidth, and ensure that the VPN tunnel terminator (e.g. firewall) has sufficient CPU horsepower to keep up with the expected demand.
Last but not least is the additional bandwidth created by large numbers of users running video conferences across the enterprise network instead of using the telephone. A typical VoIP call (assuming G.729 codec) uses 24Kbps to 35Kbps of WAN bandwidth. A video conferencing call can easily use 384Kbps or as much as 2 Mbps for that same call. The numbers add up quickly.
Video conferencing needs to be high in the QoS stack to ensure the timely delivery of video packets, thus maintaining high quality audio and video reproduction. The high bandwidth demand of desktop video is causing some enterprises to deploy a second class of service below the class used by room-based and telepresence systems. This second class of service helps them manage the growth of desktop video without impacting mission critical legacy installations.
Desktop video conferencing can grow quickly within an organization once users find out how inexpensive it is to put on the PC and how much value it adds to the collaborative experience. Enterprise IT teams need to anticipate these challenges to ensure that users will continue to be delighted with their new desktop video conferencing service.
About the author: John R. Bartlett is a principal consultant at NetForecast, where he focuses on network support for voice and video conferencing. NetForecast provides consulting to enterprises and networking equipment vendors on application performance issues and convergence of voice and video conferencing on the IP network. John has 32 years of experience in the semiconductor, computer and telecommunications, and has been consulting since 1996. John can be reached at firstname.lastname@example.org.