摘要
调度算法是多核处理器操作系统设计的核心所在,如何合理地将众多任务分配给不同的资源以使整个多核处理器系统达到最佳性能,是调度要解决的首要问题。在经典多处理器任务调度算法———ONERT算法的基础上,提出一种基于任务紧迫度的改进算法,较好地弥补了ONERT算法在处理器轻载时负载不均衡的不足。通过python程序实现了该算法,并通过实验验证了该方法在处理器轻载时实现比ONERT算法更好的负载均衡。
Scheduling algorithm is the key point of the multiprocessor operating system design.The primary problem for scheduling is to assign the many tasks to different resources so that the entire multiprocessor system can achieve the best performance.Based on the classical offline scheduling algorithm ONERT,an improved algorithm based on the urgency of tasks is presented.This algorithm can make up the load imbalance problem caused by ONERT when the processor is light loaded.Python is used to implement the algorithm,and the experiments indicate that the improved algorithm achieve better performance of load balancing than ONERT.
出处
《电子测量技术》
2012年第9期45-48,共4页
Electronic Measurement Technology
