详细信息
A method of achieving time information of communication and computation tasks from list scheduling algorithm ( EI收录)
文献类型:期刊文献
英文题名:A method of achieving time information of communication and computation tasks from list scheduling algorithm
作者:Wu, Huixin[1]; Wang, Bingzheng[2]; Wang, Yingfeng[3]
第一作者:Wu, Huixin
通讯作者:Wu, H.
机构:[1] Department of Information Engineering, North China University of Water Conservancy and Electric Power, Zhengzhou 450011, China; [2] School of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China; [3] College of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou 450002, China
第一机构:Department of Information Engineering, North China University of Water Conservancy and Electric Power, Zhengzhou 450011, China
年份:2013
卷号:9
期号:10
起止页码:3865-3872
外文期刊名:Journal of Computational Information Systems
收录:EI(收录号:20132416422416);Scopus(收录号:2-s2.0-84878748385)
语种:英文
外文关键词:Energy efficiency
摘要:In many researches on energy efficient and performance improved scheduling algorithm design for applications including computation tasks with precedence constraints, list scheduling algorithm is widely taken as comparison object due to its popularity. In order to avoid a lot of programming work, we can use existing tool to get the scheduling results from list scheduling. In TORSCHE, the corresponding function is given to get scheduling results generated from list scheduling algorithm. However, we can get the scheduling results of computation tasks without considering communication tasks according to TORSCHE. Therefore, we propose a method of achieving start time and finish time of communication tasks and computation tasks using list scheduling algorithm based on TORSCHE. This method utilizes dependency relation between computation tasks and execution time of communication tasks considering two cases: with and without competition among communication tasks. We experiment our method with random task graphs on multi-core systems with two and three processor cores, respectively. Simulation results show that our method are very effective in getting start time and finish time of computation tasks and communication tasks. This method avoids the programming work of computation task mapped to processor cores. ? 2013 by Binary Information Press.
参考文献:
正在载入数据...