|International Computer Science and Engineering Society (ICSES)|
International Transactions on Evolutionary and Metaheuristic Algorithms
Vol. 1, No. 1, Nov. 2015
A Novel Job Scheduler for Computational Grid Using Simulated Annealing Heuristic | Original Paper
Hamid Saadi a,, AbdolHamid MomenZadeh b, Ehsan PourAliAkbar c, et al. d
Highlights and Novelties
Computational Grids enable the coordinated and aggregated use of geographically distributed resources for solving large-scale problems in science, engineering, and commerce. Achieving high performance in a grid system requires effective resource scheduling. The heterogeneous and dynamic nature of the grid, as well as the differing demands of applications running on the grid makes grid scheduling complicated. Also, the execution cost, besides the completion time, has become the great concern to the grid users. Many of grid scheduling systems optimize completion time and execution cost separately. In this paper, a novel scheduling algorithm based on simulated annealing heuristic which considers both the completion time and execution cost is introduced. The proposed model applies a weighted objective function that takes into account both the completion time and execution cost of the tasks. The results obtained from our algorithm have been compared with several algorithms such as random, best of N random and climb search algorithm according to the criteria of completion time and execution cost. We show that the proposed SA scheduler produces a comparatively better result in the case of both time and cost optimization.
© Copyright was transferred to International Computer Science and Engineering Society (ICSES) by all the Authors.
Cite this manuscript as
Hamid Saadi, AbdolHamid MomenZadeh, Ehsan PourAliAkbar, et al., "A Novel Job Scheduler for Computational Grid Using Simulated Annealing Heuristic," International Transactions on Evolutionary and Metaheuristic Algorithms (ITEMA), vol. 1, no. 1, pp. 6-13, Nov. 2015.
For External Scientific Databeses
Written by: Admin | Link ...