期刊文献+

适用于多核环境的混合Cache一致性协议

Hybrid Cache Coherence Protocol Suitable for Multi-core Environment
下载PDF
导出
摘要 提出一种适用于多核环境的混合Cache一致性协议。该协议采用混合值传播策略,引入小容量目录D-Cache,克服传统监听一致性协议发送数据请求时盲目广播的缺点,通过数据块状态的扩展,有效避免乒乓现象的发生。仿真实验结果表明,该协议能减少测试程序的运行时间,降低多核处理器私有L1 Cache的失效率,提高系统性能。 This paper presents a hybrid Cache coherence protocol suitable for multi-core environment. It uses the hybrid value propagation police, introduces a small capacity directory, D-Cache. The protocol overcomes the shortcoming of traditional coherence protocol that data request is undifferentiated broadcasted. It extends data block state so that eliminates ping-pang phenomenon. Simulation experimental results show that the protocol reduces test workload runtime and L1 Cache miss rate, improves the system performance.
出处 《计算机工程》 CAS CSCD 北大核心 2011年第24期284-286,共3页 Computer Engineering
基金 国家自然科学基金资助项目(60873037 60873138)
关键词 CACHE一致性协议 多核环境 目录 数据块状态 Cache失效率 Cache coherence protocol multi-core environment directory data block state Cache miss rate
  • 相关文献

参考文献7

  • 1薛燕,樊晓桠,李瑛.多处理机系统中数据Cache的一种优化设计[J].微电子学与计算机,2004,21(12):191-194. 被引量:6
  • 2Meng Jiayuan, Skadron K. Avoiding Cache Thrashing due to Private Data Placement in Last-level Cache for Manycore Scaling[C] //Proc. of IEEE International Conference on Computer Design. Lake Tahoe, California, USA: [s. n.] , 2009.
  • 3黄安文,张民选.多核处理器Cache一致性协议关键技术研究[J].计算机工程与科学,2009,31(A01):104-108. 被引量:5
  • 4王铮,张毅.改进的基于目录的Cache一致性协议[J].计算机工程,2009,35(1):87-89. 被引量:1
  • 5Zhao Hongzhou, Shriraman A, Dwarkadas S. SPACE: Sharing Pattern-based Directory Coherence for Multicore Scalability[C] // Proc. of the 19th International Conference on Parallel Archi- tectures and Compilation Techniques. Vienna, Austria: [s. n.] , 2010.
  • 6Marty M R. Cache Coherence Techniques for Multicore Proce- ssors[D]. Madison, USA: University of Wisconsin-Madison, 2008.
  • 7Woo S C, Ohara M, Torrie E, et al. The SPLASH-2 Programs: Characterization and Methodological Considerations[C] //Proc. of the 22nd Annual International Symposium on Computer Archi- tecture. Santa Margherita Ligure, Italy: [s. n.] , 1995.

二级参考文献24

  • 1李瑛,高德远,张盛兵,樊晓桠.32位RISC中存储管理单元的设计[J].西北工业大学学报,2004,22(3):365-369. 被引量:5
  • 2Hennessy J L, Patterson D A. Computer Architecture: A Quantitative Approach[M]. 3rd ed. San Francisco: Morgan Kaufmann, 2004.
  • 3Enright Jerger N D. Chip Multiprocessor Coherence and Interconnect System Design[D]. University of Wisconsin-Madison, 2008.
  • 4Martin Milo M K. Token Coherence[D]. University of Wisconsin-Madison, 2003.
  • 5Kelteher C N, McGrath K J, Ahmed A, et al. The AMD Opteron Processor for Multiproeessor Servers[J]. IEEE Micro, 2003, 23(2):66-76.
  • 6Atoofian E, Baniasadi A. A Power-Aware Prediction-Based Cache Coherence Protocol for Chip Multiprocessors[C]// Proc of 21st Int'l Parallel and Distributed Processing Symp, 2007 : 1-8.
  • 7Barroso L A, Gharachorloo K, McNarnara R, et al. Piranha: A Scalable Architecture Based on Single-chip Multiprocessing[C]//Proc of 27th Int'l Syrup on Computer Architecture, 2000:12-14.
  • 8Ros A, Acaeio M E, Gareia J M. DiCo-CMP: Efficient Cache Coherency in Tiled CMP Arehitectures[C]//Proc of the 22nd Int'l Parallel and Distributed Processing Syrup, 2008.
  • 9Martin Milo M K, Hill, M D, Wood D A. Token Coherence: Decoupling Performance and Correctness[C]//Proc of Int'l Symp on Computer Architecture, 2003:182-193.
  • 10Strauss K, Shen X, Torrellas J. Uneorq: Unconstrained Snoop Request Delivery in Embedded-Ring Mulfiprocessors[C]//Proc of the 40th IEEE/ACM Int'l Symp on Microarchitecture, 2007:327-339.

共引文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部