摘要
针对高性能的灵巧同构系统计算环境特点,提出了一种多处理机的任务静态调度算法。该算法结合遗传算法和启发式列表调度算法,具有较短的调度时间;算法考察了多处理机调度中的任务间前趋关系、通信关系以及任务存储量要求,具有实用意义。仿真结果表明,与标准算法和CP/MISF(critical path/most immediate successors first)算法比较,所提出的算法能够得到较优的调度结果。
A static scheduling algorithm which combines the genetic algorithm and heuristics list scheduling algorithm for the smart homogeneous multiprocessor system is presented. This algorithm has short scheduling time and the practicability, thus it can be applied to the task allocation problem subject to task precedence, memory requirements and interprocessor communication costs. Compared with the standard algorithm and CP/ MISF, the proposed method is provided with a best scheduling result.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2007年第2期285-290,共6页
Systems Engineering and Electronics
基金
"十五"国防重点预研项目资助课题
关键词
多处理机
任务分配
任务调度
遗传算法
multiprocessor
task matching
task scheduling
genetic algorithm
