It’s the age-old solution to a never-ending problem. As the bandwidth of the network is limited it slows almost to a grinding halt, the shout goes out –“increase the bandwidth”. But this is simply sticking a plaster over a festering wound that just refuses to heal.
Part of the problem is a huge number of CIOs – in fact 87 per cent in a recent report – have no idea what applications are traversing their network, or how they are performing.
That’s a scary statistic, but not really all that surprising. If CIOs don’t know what’s operating on their network, how on earth do they expect to manage it to provide optimum performance to end users?
It’s vital to recognise all the applications on the network and see how they were performing. This allows you to prioritise each application across the network by putting it into the appropriate network Class of Service (CoS). Mission critical or delay sensitive traffic such as Voice and Video could get higher priority over ‘bursty’ less delay-sensitive applications like email and internet. You could even block non-business applications such as bit-torrents for example.
With hierarchical Quality of Service (QoS) you can also allow these ‘bursty’ applications to use capacity that is not being used by the higher priority applications, so this capacity is not being wasted if it is not being used. This means you can right-size your network and at the same time ensure the end users get the performance they require. That’s visibility and control.
Having solved this problem, it’s much easier to move to a converged network environment to further make better use of available network capacity. Traditional non-converged networks are great bandwidth wasters. Networks should be designed to cater for peaks in demand. We design networks to cater for the peaks in demand. If you have a separate network for your voice, data and internet, then potentially you are paying for unused capacity three times over! Some may argue otherwise, but companies should scrap this. A converged network with the with appropriate traffic management mechanisms allows you to maintain application performance without the cost of unused capacity.
