论文部分内容阅读
多任务并行执行是发挥多核处理器效能的常用手段,对多任务进行有效调度从而避免资源使用瓶颈是多核处理器任务调度的一个重要课题.本文针对多核处理器片上存储资源的使用提出一个访存负载平衡调度机制,通过刻画任务访存负载、操作系统内核层性能检测,实现了集成的计算和访存负载协同平衡系统.该调度系统在区分任务访存特征的基础上,可以在操作系统内核层组织任务执行.通过对测试程序集合的实验验证,并与其它两种调度算法进行比较,相对于操作系统缺省的计算负载平衡机制,访存负载平衡调度系统可以提高多核处理器任务吞吐量达8.2%,并且具有较低的系统实现开销.
Multi-task parallel execution is a common method to exert the performance of multi-core processors.It is an important task to schedule multi-core processors to effectively schedule multi-tasking so as to avoid the bottleneck of resource utilization.In this paper, we propose a memory access for multi-core processor on-chip storage resources Load balancing and scheduling mechanism to realize the integrated computing and memory access load balancing system by characterizing task memory load and operating system kernel layer performance.The scheduling system based on distinguishing task memory characteristics can be used in operating system kernel Layer organization task execution.Through the experimental verification of the test program set and comparison with the other two scheduling algorithms, compared with the default computing load balancing mechanism of the operating system, the memory access load balancing scheduling system can improve the multi-core processor task throughput Up to 8.2%, with lower system implementation overhead.