基于疯狂自适应樽海鞘群优化算法的异构多核任务调度
Heterogeneous Multi-core Task Scheduling Based on Crazy Adaptive Bottle Ascidians Group Algorithm
程小辉 1刘天承1
作者信息
- 1. 桂林理工大学信息科学与工程学院 桂林 541006;桂林理工大学广西嵌入式技术与智能系统重点实验室 桂林 541006
- 折叠
摘要
为了解决当前异构多核环境下的任务调度效率不能满足应用程序的多样性要求的问题,论文基于疯狂自适应的樽海鞘群优化算法(Crazy and Adaptive Salp Swarm Algorithm,CASSA),提出一种异构多核处理器任务调度算法.该算法以缩短全部任务的完成时间为目标,根据任务优先权规则设计任务分配的编码方案,利用CASSA算法中领导者的全局搜索能力和追随者的局部搜索能力,使CASSA算法在异构多核任务调度问题上有更高的收敛效率和更高质量的解.实验表明,CASSA算法的性能优良,最优解的质量高,在异构多核处理器任务调度领域中具有良好的研究意义.
Abstract
In order to solve the problem of multi-core adaptive scheduling in heterogeneous environment,this paper proposes a multi-core algorithm based on swarm optimization.The algorithm aims to shorten the completion time of all tasks,designs the cod-ing scheme of task allocation according to the task priority rules,and makes use of the global search ability of the leader and the lo-cal search ability of the follower in cassa algorithm,so that cassa algorithm has higher convergence efficiency and higher quality so-lution in heterogeneous multi-core task scheduling problem.Experiments show that cassa algorithm has excellent performance and high quality of optimal solution.It has good research significance in the field of heterogeneous multi-core processor task scheduling.
关键词
异构多核处理器/任务调度/疯狂自适应的樽海鞘群优化算法Key words
heterogeneous multi-core processor/task scheduling/crazy and adaptive salp swarm algorithm引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据