IFIP TC6 / WG6.2 and WG6.7 2nd International Conference on Network Control and Engineering for QoS, Security and Mobility, Net-Con 2003, Muscat, Umman, 13 - 15 Ekim 2003, cilt.133, ss.14-25
In this paper, we consider the problem of scheduling on a multi-stage multi-server architecture in computer centers to enhance quality of service through minimizing the average completion time of a set of requests. This problem is known tobe an NP-hard combinatorial problem. We show that the problem can modeled as a flexible flowshop problem for which few heuristics exist in the flowshop scheduling literature. We propose a new heuristic for this problem that consists of three phases. Simulated Annealing is used in the frrst phase of the heuristic followed by a greedy algorithm phase. The last phase is a local pair-wise exchange procedure. An extensive computational experiment has been conducted to compare the existing heuristics with each other for the first time. Moreover, the proposed heuristic is compared with the existing heuristics. The results indicate that the proposed heuristic significantly outperforms the existing ones. © 2003 by Springer Science+Business Media New York.