A TS-GATS Based Approach for Scheduling Data-intensive Applications in Data Grids
Data Grid technology provides services like accessing and sharing physically distributed resources for distributed scientists. In Data Grid environment, large-scale data-intensive applications which are com posed of a set of independent tasks each of which require massive size of distributed data sets that may each be replicated on different resources can be executed on different resources. In order to reduce the whole completion time of the application and improve the performance of the grid, appropriate computational resources should be selected to execute the tasks and appropriate storage resources to get the files required by the tasks. The problem can be broken into two sub-problems: one, to select storage resources and two, to assign the tasks to the computing resources. This paper presents an algorithm based on Tabu Search and Genetic Algorithm to schedule the tasks of the application. The proposed scheduler takes advantage of both Tabu Search and Genetic Algorithm. Finally, we evaluate the proposed approach comparing with other algo rithms. Simulation results show that the proposed approach can be a competitive choice for scheduling large Data Grid applications which is targeted on reducing the makespan.
Dan Liu Kenli Li Xiaoyong Tang Edwin H.M. Sha
Department of Computer Science Hunan University Changsha 410082,China
国际会议
南宁
英文
30-53
2009-12-04（万方平台首次上网日期，不代表论文的发表时间）