Unified Communications and the Network Danger

Unified Communications (or UC) describes an integrated, real-time type of communication. It's everything from video conferencing to instant messaging to telephony. Examples include things like Skype, Microsoft Lync, or Google Voice.

This technology ecosystem allows companies, both small and large, to communicate across vast distances as though meeting face-to-face. For enterprises, UC has long been on the radar. Orange Business Services reported in 2009 that 15 per cent of multinational corporations were trialling or evaluating a UC system. For SMBs, studies (such as this one by Jim Metzler) have shown that UC results in increases in productivity and large cost savings. Throughout my professional career, I've increasingly seen offices adopting UC.

There have been numerous articles written on the amazing, and very real, potential of UC. Far less is said about the risks of UC adoption. One major risk exists in the impact UC traffic flows have on the Wide Area Networks (WANs), which, in turn, impacts the performance of other business critical systems.

Let me explain this using one of my favourite metaphors: think of your networks as roads. Networks are what connect your office to other branches of the office, or to the cloud, or to your home. UC travels across these network 'roads'. It's real traffic albeit in bytes.

In the same way that the number of cars on a given road is rarely constant, so the amount of traffic from UC varies as it moves along the network. UC flows are made up of video and voice, which is dynamic traffic. This means traffic levels grow or subside based on user demand. In turn, this can result in UC applications impacting the performance of other business-critical traffic during high demand.

The network 'roads' that UC travel across have become increasingly clogged for many companies in recent years. Employees are now using company networks for their own recreational activities. This can be very, very expensive for your business. If there are traffic jams, then UC can falter or fail – which results in the ever-unpleasant experience of delay or jitter.

There are two possible solutions: you can purchase more bandwidth; and/or you can use Application Performance Guarantee (APG) solutions, which control and dynamically guarantee the performance of critical across networks.

Adding more bandwidth isn't the complete and only answer, as more bandwidth is rarely enough. Applications are bandwidth hungry. The more bandwidth you have more bandwidth they will consume. They will systematically cannibalise the successful performance of critical applications. Additionally, the added bandwidth will struggle to guarantee critical application performance.

In general, application traffic tries to use up all the available bandwidth. Simply increasing it is like filling a bottomless pit: expensive and rarely enough to satisfy ever-growing usage demands. The additional traffic may also hinder the performance of the business-critical and often resource-thrifty applications.

Another issue lies in the inability of additional bandwidth to respond appropriately to the variable nature of UC traffic. Perhaps during the call, an employee decides to share a large file. Another streams a video from an online host. Suddenly, 100 per cent of the company's bandwidth is used, with several other users attempting to make VoIP calls or update SaaS CRM Systems. The bandwidth is already at capacity. Something will have to give somewhere, but the network doesn't know where the 'give' should occur. Nor does it offer the ability to execute network control.

Finally, additional bandwidth doesn't allow for efficient troubleshooting. When issues do occur as traffic moves across the network, having a larger 'road' won't allow companies to see where, when, and why problems happen.

Basically, you need solutions that will allow you to control and guarantee the performance of key applications as they use the network. These solutions must work anytime and everywhere.

This type of monitoring and controlling is known as an "Application Performance Guarantee" (APG) solution. APG solutions allow you to see how each application is using the network, where, when and why. You can then guarantee the performance of top, business-critical applications, optimise end-user experience, manage global network resources, and reduce costs.

Let's go back to the car example.

Networks are roads. Applications are cars. An APG solution in this case might be viewed as a police officer. It can direct cars into appropriate queues. It can slow cars or speed cars depending on their priority. This allows the vehicles to get to their destination in a timely fashion – regardless of the amount of traffic on the 'road'. It is very much about ensuring network resources are allocated in the most efficient way possible.

Such solutions provide guaranteed performance and full visibility. This means the impact of UC deployment can be fully understood and intelligently routed across the network.

If I were to summarise all the above, I'd say this: UC offers sizable benefits and will continue to be adopted. It's happening everywhere. Yet it threatens network performance and in turn other business-critical applications. It is important to consider this before UC goes live to ensure smooth UC deployment.

Béatrice Piquer-Durand has over 18 years' experience in the IT industry, where she has been focused on the development of marketing strategies for start-up and IT companies. Before joining Ipanema Technologies in 2001 to secure the sales success of the company and to create the worldwide marketing team, Béatrice worked as marketing director for several companies including Thomson Multimedia (Technicolor), Catalliances (Prodware) and Interface Data.