A Novel Processor Allocation Policy for Task Scheduling Algorithms on Distributed Heterogeneous Computing System

초록

The performance of DHCS depends on an algorithm which schedules input DAG. General scheduling algorithm consists of task prioritizing phase and processor allocation phase. Among various processor allocation methods, most scheduling algorithms use either insertion or duplication method. In this paper, we propose a new processor allocation policy, called HAP, which uses both insertion and duplication method. The performance of HAP has been compared by applying it to task prioritizing phase of conventional scheduling algorithms. The comparison results show that HAP provides better performance than algorithms with conventional processor allocation polices.