Learning Minimum Delay Paths in Service Overlay Networks

Free registration required

Executive Summary

The authors propose a novel approach using active probing and learning techniques to track minimum delay paths for real-time applications in service overlay networks. Stochastic automata are used to probe paths in a decentralized, scalable manner. They propose four variations on active probing and learning strategies. It can be proved that their approach converges to the user equilibrium for minimum delay routing. The performance of these strategies is studied via fluid simulations of a model of AT&Ts backbone network. The simulation results show that the proposed strategies converge to the minimum delay paths rapidly.

  • Format: PDF
  • Size: 292.05 KB