In this paper we present a new algorithm for the off-line scheduling of parallel and independent jobs on a massively parallel processor system. We also introduce a machine model which is based on existing multiprocessors and accounts for the penalty of preemption. It is shown that the new algorithm achieves a small approximation factor for both weighted completion time and makespan scheduling. To fine tune the algorithm a fairly simple numerical optimization problem is derived. This way different preemption penalties can be considered when determining the approximation factor. Finally, we compare the generated schedule with non-preemptive schedules for the same problem.
CITATION STYLE
Schwiegelshohn, U. (1996). Preemptive weighted completion time scheduling of parallel jobs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1136, pp. 39–51). Springer Verlag. https://doi.org/10.1007/3-540-61680-2_45
Mendeley helps you to discover research relevant for your work.