The New Task Scheduling Algorithm in Computing Grid Environment

• Main Authors: Jen-Jom Chen, 陳建仲
• Other Authors: rang shung chang
• Format: Others
• Language: en_US
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/44572384247243960580

碩士 === 國立東華大學 === 資訊工程學系 === 95 === In the past, supercomputer is the only choice for complex computations in science, engineering and business computing. With the speed-up of personal computers and networks, the superiority of supercomputer is no longer ture. The grid technology, which connects a set of personal computers with high speed networks, can provide more powerful computing than the best supercomputer today. The grid environment can be classified into three types: data grid, computing grid, and access grid. We focus on computing grid in this thesis. In computing grid, the task scheduling to discover resources for user’s requirements is important. In general, task scheduler assigns tasks to proper resource node for execution, and the resource nodes with better performance would be assigned first. When task loading is heavy and all resource nodes with better performance are assigned, other tasks might be assigned to the resource nodes with normal or worse performance. If a task is heavy loaded but assigned to the resource node without proper performance, the overall execution time will increase obviously. To solve this problem, a task scheduling algorithm which searches the proper resource for task is proposed in this thesis. Moreover, some scheduling algorithms, such as FPLTF (Fastest Processor to Largest Task First), work well only in static environment. In this thesis, simulation results show the proposed algorithm works well in both static and dynamic environments.