Adaptive traffic distribution in optical burst switching networks

Abstract

Optical burst switching (OBS) is a promising technology for realizing next generation high speed backbone optical networks. Burst contention is a major problem in OBS networks due to the lack optical memories at the core nodes. Effective multi-path traffic distribution schemes are important methods in reducing burst contention in OBS networks. This research focuses on the problem of adaptive traffic distribution in OBS networks. Firstly, we develop an adaptive proportional routing scheme which attempts to route the incoming traffic in OBS networks at the flow level. It avoids the problem of packet out-of-sequence arrival which is a common problem in existing adaptive traffic distribution schemes. Next, we propose a multi-path traffic routing scheme in OBS networks based on the well-known gradient projection optimization framework. The proposed scheme performs well in minimizing network congestion and exhibits good routing stability. The performance of all the proposed methods is evaluated through simulation experiments.