Get Ready for Video (Finally!)
For years we've heard that the age of ubiquitous video conferencing was just around the corner, but now a perfect storm has formed, driving demand for video conferencing within the enterprise, and causing potential for chaos for network managers.
For years we've heard that the age of ubiquitous video conferencing was just around the corner, but now a perfect storm has formed, driving demand for video conferencing within the enterprise, and causing potential for chaos for network managers.IP video conferencing adoption is growing, with more then half of the 400+ companies participating in the Nemertes benchmark Unified Communications and Collaboration adopting desktop and room-based video conferencing, while another 38% are deploying immersive telepresence systems from vendors including Cisco, Polycom, Tandberg, Teliris, and HP. These numbers are up substantially from our last round of interviews in late 2006-early 2007. Why?
The following trends have converged at the right time, both for users of video as well as vendors/service providers offering video solutions:
So while these developments come as welcome news for video system vendors as well as virtual workers, enterprise IT managers are left struggling with two challenges: how to decide on the "right" approach for video, and how to deliver a network infrastructure capable of supporting the most demanding of all applications, one that requires not only low latency but also large amounts of bandwidth.
Determining the "right" approach is complicated by varying marketing terminology that creates confusion between different types of systems. For the sake of providing a reference point, we define video conferencing systems as falling into one of the following three buckets:
Video conferencing systems may incorporate one or more of the above approaches, ranging from those entirely based on desktop video to those that enable participation by anyone using any of the three defined interfaces.
Deciding on the "right" approach needs to take into account a variety of factors such as the number of users, the quality desired, available bandwidth, cost, current systems & software, and the need for platform ubiquity (e.g. supporting Linux, Mac and/or Windows). Organizations may be able to leverage existing investments in video or telephony systems and simply append video, while those just starting out will need to decide if they need a dedicated video solution or if they would benefit more from a video product that's integrated with other UC applications such as voice and/or web conferencing.
From the network perspective video, especially desktop video, HD, and telepresence, can wreak havoc on network designs and capacities. Desktop video is often a peer-to-peer application (though aggregation points are typically required for multi-party video conferences). With bandwidth requirements usually ranging from 128 kbps up to 384 kbps, a large adoption of desktop video can generate a great deal of bandwidth demand, especially troublesome for organizations still relying on hub & spoke network designs.
HD and telepresence have their own network demands that can be spelled out in one word - bandwidth (and lots of it). HD systems generally require at least 1 Mbps per screen, with some telepresence systems requiring dedicated bandwidth of 45 Mbps per room for a multi-screen room. With the absence of switched bandwidth services, enterprises will have to pay for the additional bandwidth to support HD or telepresence even if it isn't being used.
The bottom line for network managers is that video is coming, if it isn't here already. Get out in front of video adoption plans within your organization, develop a strategy, and plan for bandwidth needs as required.