基于本体的多核系统缓存竞争研究
Cache Contention on Multicore Systems: An Ontology-based Approach
摘要:
多核处理器已被证明是台式机和服务器系统的正确选择,因为它可以以可接受的预算支出支持高性能。在这项工作中,我们比较了几个作品在缓存竞争,并发现这些作品已经确定了几种技术的缓存竞争以外的缓存大小,包括前端总线,内存控制器和预取硬件。我们发现,分布式强度在线(DIO)是一个非常有前途的缓存竞争算法,因为它可以实现高达2%的最佳技术。此外,我们提出了一个新的框架缓存竞争的基础上资源本体。其中本体实例将用于不同进程之间的通信,而不是基于硬件的抓取调度。
英文摘要:
Multicore processors have proved to be the right choice for both desktop and server systems because it can support high performance with an acceptable budget expenditure. In this work, we have compared several works in cache contention and found that such works have identified several techniques for cache contention other than cache size including FSB, Memory Controller and prefetching hardware. We found that Distributed Intensity Online (DIO) is a very promising cache contention algorithm since it can achieve up to 2% from the optimal technique. Moreover, we propose a new framework for cache contention based on resource ontologies. In which ontologies instances will be used for communication between diverse processes instead of grasping schedules based on hardware.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。