【作者】 赵长林； 辛阳； 杨义先；

【Author】 Changlin ZHAO,Yang XIN,Yixian YANG 1.Information Security Center,Beijing University of Posts and Telecommunications beijing,100876 2.National Engineering Laboratory for Disaster Backup and Recovery,Beijing University of Posts and Telecommunications beijing,100876 3.Beijing Safe-Code Technology Co.,Ltd.Beijing,100876

【机构】 北京邮电大学信息安全中心； 北京邮电大学备灾技术国家工程实验室； 北京安码科技有限公司；

【摘要】 cache一致性共享存储器体系结构应用得越来越广泛,虽然他们的并行性和峰值性能也在不断地提高,但是由于内存栅栏和cache行错误共享的存在,其效率往往是令人失望的.特别是在大流量数据流处理时,锁的使用加剧了不同任务间的竞争,降低了并行处理带来的性能改进.为此提出一种新型的算法,它是基于对Lamport单生产者/单消费者环形缓冲队列的改进,实现了无阻塞lock-free/fence-free同步机制,避免了错误共享引起的cache行抖动带来的性能损失.更多还原

【Abstract】 Cache-Coherent shared-memory multi-core architectures are becoming increasingly popular.While their parallelism and peak performance is ever increasing,their efficiency is often disappointing due to memory fence overheads and false sharing.Especially in the large flow of data stream processing,the use of locks exacerbated competition between different threads,reducing performance improvements brought about by the parallel processing.In this paper,For this reason,a new algorithm is proposed,which is based upon Lamport’s circular buffer queue,achieving non-blocking lock-free/fence-free synchronization mechanisms,and avoidomg the performance loss caused by cache-line thrashing due to false sharing.更多还原