Grid ia an environment that integrates calculation and resources, It canfully absorb all kinds of computing resources and turn them into a computingcapability which is easily available, reliable, standardized and economical.In this way, we can share the resources thoroughly. Scheduling is an importantcomponent element in sharing resources effectively. Because of the charactersof dynamic,distributed and different requirements to resources asked byapplications running in the Grid Systems, it makes the task-scheduling verycomplex. This dissertation' tasks are as follows:First, it goes deeply into the studty of the grid scheduling' theory andfeatures. Also, it has a research on the Scheduler' organizational structureand gives some suggestions. Together, it analysis the determinants of gridscheduleing.Second, after researching on the classic algorithm, this dissertationsuggests a grid task scheduling based on a mapping strategy Through detailedanalysis of the new strategy, this dissertation gives a algorithm and comparesit with other algorithms based on other strategies which are classical inexperiments.Third, Genetic Algorithm is a kind of holistic searching algorithm withdirectionality. This paper designs a grid task scheduling based on areformative Genetic Algorithm. In order to get the best result with a fastconvergent speed, this paper conducts an indepth theoretical analysis andexperimental research.Finally, all algorithms are simulated in the Grid Simulator. The resultsshows that the algorithms are effective and the experiments have achieve theexpected goal. |