Monitoring the effects of high bandwidth videoconferencing on the network

Given its many promises of increased productivity, videoconferencing continues to gain popularity in the workplace, both in enterprise-grade systems and simple desktop video apps like Skype. But these benefits come at a cost, with networks becoming clogged with high-bandwidth video traffic.

A survey run by SolarWinds confirms predictions from last October’s Oracle OpenWorld trade show that “video will be the platform for all forms of communication in IT as we go forward”.

The soothsayer was no less than John Chambers, Cisco’s chief executive.

He also predicted that by 2013, 91% of all network traffic will be voice/video streams.

While that might be expected from the head of a company with a vested interest in proliferating video delivery, the more recent SolarWinds survey confirms this trend. The survey found that:

• 89% of respondents have deployed some level of video service within their network
• assessment of network readiness to support video traffic is viewed as critical
• 75% of respondents indicated that video monitoring and troubleshooting are either ‘Important’ or a ‘Must Have’ in their network monitoring system

So, how does video impact the network and how do sysadmins go about ensuring both network performance and video quality? Don’t think for a minute that because you haven’t built-out a Cisco Telepresence or Polycom system that you are not at risk of video significantly impacting the performance of your network. The real impact likely will come from the use of desktop video.

Today, desktop video is inexpensive to deploy. High-quality web cams can be purchased for under $100, or they are already integrated into your laptop. Add free software (like Skype) or a package that may be included with another licence (like Microsoft Office Communicator) and voilà, you have deployed videoconferencing. The next thing you know is that it is going viral. Heck, I get a video call from my daughter every afternoon when she gets home from school.

So, I repeat the question: what is the impact on your network of all of this video and how do you ensure network performance and quality? I believe we simply don’t know the answer yet. Every enterprise will be different, depending on its use cases and IT infrastructure. By proactively monitoring the network, sysadmins can at least understand the current impact of video and plan for future capacity needs.

Using traditional network troubleshooting techniques such as ping or interface statistics will provide limited information that will not allow you to pinpoint the problem and location of the problem. With more advanced techniques, such as IP SLA monitoring, you can pinpoint a segment of your network that may be causing the problem (ie, a WAN link) by monitoring Real Time Protocol (RTP), TCP and IP constant bit rate (CBR) traffic on network devices. This will allow you to measure audio and video service metrics like packet loss, network jitter and round trip time.

Video and network monitoring can be accomplished using either advanced hardware appliances or network and traffic monitoring software products, each having their own advantages and disadvantages. Hardware appliances tend to have more features but usually come with the disadvantage of being more costly and harder to maintain while software-based approaches are less expensive and easier to deploy.

Regardless of the chosen approach, it is essential to select a solution that measures and monitors overall network and traffic performance to be sure of seeing the impact of video on all of network elements. Measuring video call data records (CDRs) in real time will provide information about utilisation patterns and demand trends, so administrators can start to predict capacity needs. In addition, they should look for the ability to generate synthetic flows in order to simulate the impact of various types of video communications on the network.

So, whether a video deployment is formal or viral, there’s no hiding from the fact that it will have an impact on network performance. Start monitoring now to understand better just what the impact will be.

*Sanjay Castelino is a Vice President and Market Leader at SolarWinds, IT management software provider. Castelino leads the company’s initiatives around its end-to-end IT solutions for network, security information and event management (SIEM), storage and virtualisation management. He previously held the positions of Vice President of Marketing and Business Development at NetStreams, and Vice President of Product Marketing and Management at Motive.