Worst-case schedule bounds and performance evaluation of a near-optimal algorithm for scheduling of independent tasks on multiprocessors

This paper discusses the problem of assignment of independent tasks to processors and proposes an algorithm, the Variant-Load algorithm. for balancing the workload on each of the p processing elements used for scheduling of the tasks. An analysis of the worst-case schedule bounds of the Variant-Load algorithm is presented. The performance of the proposed algorithm is compared to several of the best known comparable algorithms in the literature. It is shown that the Variant-Load algorithm performs at least as well as the best known algorithms in terms of the optimality of the prcduced schedules and is superior to these algorithms in terms of its run-time complexity.