Research On Scheduling Algorithm Based On QoS

Grid is an important information technology raised on international in recent yearswith the aim of making plentiful heterogeneous resources distributed on broad geographyshared by all users. Task scheduling is an important part of grid which using appropriatestrategy to distribute different tasks to proper resource nodes according to the informationof task. Task scheduling includes three parts: resource discovery, resource match and taskperformance. Since the grid system is heterogeneous, dynamic and the applicationsrunning on the grid needs different resources, task scheduling becomes very complex. Abad resource discovery method and task allocation strategy will increase theimplementation time of tasks and reduce the throughput of the whole grid system.Therefore, the author researched into the current common resource discovery andmanagement model and put forward a resource classification based grid resourcesdiscovery model to solve the problem of inefficiency caused by stack transfer inhierarchical model, which will improve the speed of resource finding and updating greatly.She also analyzed the excellent static and dynamic scheduling algorithms in the world,especially the classical Min-min algorithm and QoS guided Min-min algorithm.Considering the influence of different requirements of task on scheduling algorithm, theauthor put forward a proper algorithm to balance the charge and increase the throughputefficiently. Finally, the author used GridSim tools to make simulating experiment to testthe algorithm she proposed. The result of the simulation showed that the improvedscheduling algorithm is more efficient compared with the Min-min algorithm and QoSguided Min-min algorithm.