期刊名称：International Journal of Computer Science Issues
印刷版ISSN：1694-0784
电子版ISSN：1694-0814
出版年度：2012
卷号：9
期号：3
出版社：IJCSI Press
摘要：A grid is a heterogeneous computing and communication system that allows a group of nodes to compass a task. The process of assigning the jobs or subtasks to the nodes present in the heterogeneous grid is known as scheduling. The type of scheduling in which the subtasks are assigned to the nodes is known as Local Scheduling.A significant research problem is how to assign resources to the subtasks and order the execution of the subtasks that are matched to maximize performance criterion of a local grid system. This procedure of scheduling is called mapping or resource allocation. MidSFN adverts to Medium Subtask Fastest Node algorithm which classifies the subtasks into three tier categories, High, Middle and Low based on their priority. In MidSFN algorithm priority is assigned based on the new parameters Computational Complexity and Processing Power. The value for processing power is assigned based on the Performance Factor. The value of the Performance Factor is the product of the number of operations per cycle per processor and the number of instructions processed per second. In MidSFN algorithm the subtask of medium computational complexity and resources exhibiting medium processing power are assigned with a high priority. The subtasks are then mapped to respective processors based on the assigned priority for execution. Compared to other local scheduling algorithms, MidSFN algorithm shows efficient load balancing and better computation with effective usage of resources. The effectiveness of this algorithm is evaluated through simulation results.
关键词：Computational Grid; Local scheduling; Computational Complexity; Directed Acyclic Graph (DAG)
