Network congestion management

Congestion is an ongoing network issue for every operator, regardless of the size or access technology. Every network is designed and built within a budget, to support a specific peak throughput. The general network growth trend is an approximate doubling every 2 years. The main driver of this is video and according to data from Sandvine, Google and Facebook account for the majority of all traffic.  BitTorrent filesharing is also still a major source of traffic.

Short term bursts of traffic can hit a design limit and badly impact user experience, by introducing packetloss or latency or both.  If the steady state is below the safe working limit, it can be difficult to justify an across the board expansion, just for occasional peaks.

This is where intelligent traffic management can be very useful. The main purpose is to intelligently manage short term traffic bursts.

Providers of public networks within the UK and elsewhere are often restricted in what they can do by Net Neutrality. Roughly, this says that each traffic type must be treated equally. Operators of private / enterprise networks have no such obligation, so are much freer to manage traffic to suit their business needs.   Is Facebook related traffic a legitimate business requirement to support? TicTok? Instagram? WilliamHill? PirateBay?

For a typical person, trying to focus on a particular task, having an interruption, even if only for a few seconds can cause many minutes of lost productivity, as people have to remember where they left off, recall their train of thoughts and restart. Facebook, Instagram, WhatsApp and similar apps are a constant source of interruption and hence lost productivity.  In extremes, an interruption can lead to a mistake, causing losses and further lost productivity.  In short, reducing interruptions is critical to improving productivity.

The key first step is to know what is happening on your network:

1. How much non-business related traffic is there?
2. Am I seeing bursts hitting my capacity limits?
3. When is packetloss occurring?
4. How is latency varying throughout the day?
5. Do we have any filesharing traffic, which could lead to copyright infringement notices being served?
6. Is there gaming or gambling traffic during working hours?

Once you have a picture of typical traffic on a corporate network, it is possible to start to manage it, in order to save unnecessary costs to expand capacity and also to help improve employee productivity.

Once certain traffic types have been identified, the next step is to decide how to manage them. It might be tempting to block 100% of unwanted traffic, but some protocols will react to this and make it harder to track and employees will notice. A better approach is simply to reduce allowable bandwidth to a low rate. In some cases, it might become unusable, however complaints are unlikely if the traffic type is not legitimately business related.

A further approach is to add time based rules, so social media traffic could be allowed to increase during lunchtime and after work for example.

By prioritizing legitimate business traffic only, the performance of the network is improved and productivity increases, with fewer distractions for staff.  The cost to expand the network and internet connectivity can be reduced.

ActiveLogic enables management of traffic across the network.  Deployable on COTS hardware, ActiveLogic can react to peaks in demand in milliseconds. This allows you to run your network near capacity much more safely, in the knowledge that only important traffic is getting through.