Network Traffic Profiling services are wrapped around a packet-shaping and performance-reporting tool. This classifies and analyses network traffic, enforces policy-based bandwidth allocation, and generates reports on network utilisation and performance through the selected circuit.

In order for the platform to manage traffic, traffic must pass through the platform at a useful point in a network. Exactly where that point is depends on the customer’s network configuration. In general, traffic profiling is most beneficial at the edge of a network or at remote sites.

The platform measurement engine runs as a background process, continuously collecting time-series and histogram data. This data is stored on the platform unit’s hard drive. The measurement data is accessed via the browser interface or the ‘measure’ command in the CLI. Gen-i typically removes the appliance from your site and extracts the data offsite. Having this tool on your network has no impact on bandwidth, and minimal impact upon latency.

A typical appliance deployment is illustrated below. This diagram shows the typical placement of two traffic profiling platforms in a network. With these placements:

• The traffic profiling platforms see all traffic regardless of destination.
• The traffic profiling platform at the remote end can effectively manage any bottlenecks because the ports at the remote office are smaller than at Head Office.
• The traffic profiling platform at the remote end manages the rate control of both inbound and outbound TCP traffic.