Applying Network Policy Control to Asymmetric Traffic
To be a viable solution in modern communications networks, stateful packet-processing equipment must overcome the challenges posed by routing asymmetry and immense scale.
Routing asymmetry increases network efficiency and redundancy, but also presents serious challenges that must be overcome so that CSPs can implement critical use cases such as accurate charging, policy-based measurements, and congestion management.
The rapidly increasing volume of Internet traffic worldwide also presents issues of efficiency, extensibility, and redundancy.
Sandvine’s Policy Traffic Switch Clusters
On a per-unit level, the PTS is impressive, but perhaps the most differentiating characteristic is the capability to combine multiple PTS units into a cluster that behaves as a single enormous platform. Clustering works by combining the intersection and processing capacity of many units into one and ensuring that all packets associated with a flow, session, and subscriber are always processed by the same processor core, regardless of the interface by which they enter the cluster.
A cluster of Policy Traffic Switch units emulates the behavior of a single enormous network policy control box in order to overcome asymmetry, deliver extreme scale, and preserve redundancy; importantly, clustering achieves these goals with incredible efficiency. The PTS architecture (at a per unit level and at a cluster level) is ‘shared nothing’, which ensures linear extensibility and enables economical N:N+1 redundancy.
Importantly, a PTS cluster is managed as a single unit, so there is no added operational management complexity, and does not introduce any meaningful latency.
PTS clustering is a beautiful, elegant, and – above all – effective and efficient solution to the challenges imposed by routing asymmetry and the demands of massive scale.