计算机工程与应用Issue(10)：264-270,7.DOI:10.3778/j.issn.1002-8331.1610-0303
网络并行计算中多处理机任务调度问题研究
Study of multiprocessor task scheduling problem in network parallel computing system
摘要
Abstract
In the network parallel computing system, the problem which aims to schedule tasks with multiprocessor and multistage is common. Hence, this paper proposes the model of Multiprocessor Task Job-shop Scheduling Problem, i.e. MTJSP, a combination of Multiprocessor Task Scheduling(MTS)and Job-shop Scheduling Problem(JSP). MTJSP in which every task needs more than one stage to complete is different from the classical MTS in which there is only one stage for each task. The mathematic formulation of MTJSP is constructed and a Hybrid Particle Warm Optimization(HPSO) is designed for solving it which includes the decoding schemes, new ways for particle updating, memories for better solu-tions and function of local searching. Plenty of instances are used to measure the performance of HPSO and numerical results show that HPSO behaves very well on instances of both JSP and MTJSP.关键词
多处理机任务/作业车间调度/粒子群优化算法/局部搜索Key words
multiprocessor task/Job-shop Scheduling Problem(JSP)/Particle Swarm Optimization(PSO)/local search分类
信息技术与安全科学引用本文复制引用
王蒙,樊坤,翟亚飞,李心宁..网络并行计算中多处理机任务调度问题研究[J].计算机工程与应用,2017,(10):264-270,7.基金项目
教育部人文社科基金(No.14YJC630030) （No.14YJC630030）
国家自然科学基金(No.71502015,No.71402006) （No.71502015,No.71402006）
北京社会科学基金(No.16GLC059) （No.16GLC059）
北京高等学校青年英才计划项目(No.YETP0776). （No.YETP0776）
