Genetic algorithm based scheduler for computational grids.

Computational grids can be used to solve grand challenge problems of scientific research and to handle peak processing demands in large organizations. For managing these tasks, grids may use distributed and heterogeneous compute resources, including commodity machines and supercomputers. This thesis deals with highly scalable distributed resource management architecture for the global grid. The main component in the architecture is the grid scheduler. A scheduler must use the available resources efficiently, while satisfying competing and mutually conflicting goals. Genetic Algorithm is used to obtain the best schedule for mapping of tasks to compute-nodes. The grid workload may consist of multiple jobs, with quality-of-service constraints. Each job has tasks with arbitrary precedence constraints and arbitrary processing time. A Directed Acyclic Graph (DAG) represents each such job. The thesis presents the design, implementation and test results for a genetic based grid scheduler. It attempts to minimize make-span, idle time of the available computational resources, turn-around time and the specified deadlines by the users. The architecture is hierarchical and the scheduler is usable at either the lowest or the higher tiers. It can also be used in both the intra-grid of a large organization and in a research grid consisting of large clusters, connected through a high bandwidth dedicated network.Dept. of Computer Science. Paper copy at Leddy Library: Theses & Major Papers - Basement, West Bldg. / Call Number: Thesis2004 .A34. Source: Masters Abstracts International, Volume: 44-01, page: 0380. Thesis (M.Sc.)--University of Windsor (Canada), 2005