Optimal algorithms for the batch scheduling problem in OBS networks

Abstract

This paper introduces optimal batch scheduling algorithms for the problem of scheduling batches of bursts in optical burst switching networks. The problem is modelled as a job scheduling problem with identical machines. The consideration of previously scheduled bursts in the scheduling allows such modeling. Two optimal algorithms with polynomial time complexity are derived and evaluated. Results show that the algorithm that allows re-scheduling of previously scheduled bursts leads to preferred solutions. Moreover, an extended version of the JET reservation protocol is proposed for efficient handling of batches of bursts. Results obtained via simulation prove the superior performance of the BATCHOPT algorithm.