Research On Genetic Algorithm Of Job Scheduling In Grid Based On Auction Mechanism

Grid computing is a kind of distributed computing. But the environment of grid computing is more complex than the environment of distributed computing, so the resource scheduling policy which design on distributed computing unsuitable the environment of grid computing. For example, traditional policy unable forecast resources' dynamic charge, think over the confines of resourcesjobs, the cost of finish the jobs and some other problems in grid computing.This paper present an online reputation grid resource management model based on auction mechanism. This resource management model is based on the grid architecture for computational economy. It provides the motivation of resources contribution and share for resource providers, and attracts more better resources to join for achieving the resource optimal allocation. The resource management model guarantees maximum benefits for both transaction sides, so it contributes to the market management of grid resource and balance between supply and demand.This paper discusses the design the global scheduling strategy of grid jobs based on the Genetic Algorithm. In the algorithm, we introduce two economy factors which are time and cost. The different influential factors are set by user, then the mathematical relation between the two economy factors is deduced through researching two factors. The algorithm can achieve the optimized resource allocation strategy on user demand, thus improve the system's performance.A concrete grid environment is simulated by gridsim toolkits. The simulation environment has contained all basic factors which are needed in general grid environment. The global scheduling strategy of grid job based on Genetic Algorithm is tested on the simulator GridSim to estimate the performace of algorithm, the results show that the global scheduling strategy of grid job based on Genetic Algorithm can reduce total processing time and total cost of user job effectively, thereby the utilization ratio and performance-to-price ratio of grid system are improved.