PAPC: A simple distributed parallel computing framework for mass legacy code tasks
With the fast development of multi-core technology, future high performance computing system will be a system of many multi-core processors where parallel computing is a must. However, it is not an easy task in such an environment to run legacy code programs that are developed for machines of singlecore processors to achieve the desired speedup. To deal with the problem, in this paper we put forward a parallel computing framework named PAPC on the basis of ProActive, an open source middleware that aims to ease the development of parallel and distributed applications and that has excellent support for high performance computing. We explain in detail the design and implementation of the PAPC and conduct some experiments to evaluate it. Experimental results show that the framework PAPC is of strong scalability, good adaptability and can make full use of the high performance of multi-core.
ProActive high performance computing legacy code parallel computing multi-core
Xiaofeng Wu Jinlei Jiang Jinzhe Yang
Department of Computer Science and Technology Tsinghua National Laboratory for Information Science and Technology Tsinghua University, Beijing 100084, China
国际会议
The Ninth International Conference on Grid and Cloud Computing(第九届网格与云计算国际学术会议 GCC 2010)
南京
英文
526-532
2010-11-01（万方平台首次上网日期，不代表论文的发表时间）