Performance Analysis of Resource Selection Algorithms in Grid Computing Environment | Science Publications

Journal of Computer Science

Rhymend Uthariaraj¹  Malarvizhi Nandagopal¹ 

关键词: Performance analysis;    LARS algorithm;    computing environment;    Meta- scheduling;    Multi Criteria Resource Selection (MCRS);    Grid Information Server (GIS);    Power Aware Resource Selection (PARS);   

DOI  :  10.3844/jcssp.2011.493.498

学科分类：计算机科学（综合）

来源: Science Publications

【 摘 要 】

Problem statement: Meta- scheduling has become important due to increased number ofjobs and resources in the area of grid computing to achieve the objectives of grid users and gridresources. Approach: A variety of factors need to be considered for effective scheduling of resourcesin such environments such as total time job spend in the grid system, global and local allocationpolicies and resource utilization. A general and extensible scheduling architecture addressing theseissues is proposed. Within this architecture a Multi Criteria Resource Selection (MCRS) algorithm isdeveloped and performance of this algorithm is evaluated using simulations for a wide range ofparameters. To select a resource for job execution the proposed algorithm considers multiple criterialike processing power, workload and network bandwidth of the resource. Results: The proposedalgorithm is compared with the conventional single criteria resource selection algorithms.Conclusion: Simulation results show that in order to improve the performance of both user andresource it is important to consider multiple criteria together to select the optimal resource, rather thanconsidering them separately.

【 授权许可】

Unknown