Currently, applications are at the mercy of a network’s infrastructure for the selection of a path within a network environment where more than one path exists between a source and a destination. Too often, the network infrastructure elements are unaware of an application’s requirements, or are aware of them in only a very rudimentary way. This situation is particularly dire for collaboration applications, which often have the most stringent requirements for path characteristics including delay, jitter, and packet loss. Techniques are presented herein that move the point of control for path selection to a collaboration application through a lightweight, in-band signaling mechanism that is exposed by the application to a network’s infrastructure for appropriated and differentiated traffic routing. Aspects of the presented techniques support the use of a per-application tunneled path for traffic flows, combined with a measurement methodology for those multiple paths and a mechanism for the application-level designation of specific and differentiated traffic pathing via an upstream router, allowing an application to measure performance across multiple paths and then signal to a network which path to choose based on per-application preference and service-level agreement (SLA) criteria.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.