期刊文献+

一种基于混合存储的高效、可扩展细粒度缓存管理研究 被引量:1

Research about Efficient and Scalable Hybrid Memories at Fine-granularity Cache Management
下载PDF
导出
摘要 混合型主存储器也是DRAM构成,并把混合型主存储器作为高速缓存使用,因此可以扩展Non-volatile存储器,相比传统型的主存储器,混合型主存储器可以提供更大的存储能力.然而,对于混合型主存储器,要使其具有高性能和高可扩展性,一个关键的挑战是以一种细粒度的方式有效地管理缓存在DRAM中的数据的元数据(如Tag).基于此观察:根据DRAM高速缓存行的局部性,在芯片外相同的高速缓存行中存储数据和数据的元数据,本文通过使用一个小缓冲区缓存最近被访问的片内缓存行来减少细粒度DRAM高速缓存的开销.基于细粒度DRAM高速缓存使用的灵活性和效率,还开发了一种自适应的策略,可以选择数据迁移到DRAM高速缓时的最佳迁移粒度.在配置了512MB DRAM的混合存储系统中,使用8KB的片上高速缓存,相比传统的8MB SRAM元数据存储,即便没有考虑大SRAM元数据存储的能耗开销,实现了不超过6%的性能提升,节约了18%左右的能效. Hybrid main memories are composed of DRAM which can provide much larger storage capacity than traditional main mem- ories used as a cache to scalable non-volatile memories, such as phase-change memory. However, for Hybrid main memories with high performance and scalability, a key challenge is to effectively manage the metadata { e. g. , tags } for data cached in DRAM in a fine- granularity. Based on this observation: storing metadata on off-chip cache line in the same row as their data corresponding to the metadata exploits DRAM row buffer locality, this paper reduce the overhead of fine-grained DRAM cache by using a small buffer cache chip cache line has recently been accessed. We also develop an adaptive policy to choose the best granularity when migrating data into DRAM. On a hybrid memory with a 512MB DRAM cache, our proposal using an 8KB on-chip buffer can achieve within 6 % of the performance of, and 18 % better energy efficiency than, a conventional 8MB SRAM metadata store, even when the energy overhead due to large SRAM metadata storage is not considered.
作者 姜国松
出处 《小型微型计算机系统》 CSCD 北大核心 2014年第2期334-338,共5页 Journal of Chinese Computer Systems
基金 湖北省自然科学基金项目(2011CDC078)资助
关键词 高速缓存 标签-存储 非易失性存储器 混合型主存 cache tag memory non-volatile memories hybrid main memories
  • 相关文献

参考文献16

  • 1Lee B C, Ipek E, Mutlu O, et al. Arehitecting phase change memory as a scalable DRAM alternative[ C]. Proceedings of The 36th Inter- national Symposium on Computer Architecture,2009:171-182. 被引量:1
  • 2Qureshi M K, Srinivasan V, Rivers J A. Scalable high performance main memory system using phase-change memory technology [ C ]. Proceedings of the 36th International Symposium on Computer Ar- chitecture ,2009 : 101-112. 被引量:1
  • 3Dong X,Xie Y,Muralimanohar N ,et al. Simple but effective heter- ogeneous main memory with on-chip memory controller support [ C ]. Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis, 2010:773-778. 被引量:1
  • 4Zhao L, Iyer R, Illikkal R, et al. Exploring DRAM cache architec- tures for CMP server platforms[ C]. Proceedings of 2007 1EEE In- ternational Conference on Computer Design ,2007:253-259. 被引量:1
  • 5Loh G, Hill M D. Efficiently enabling conventional block sizes for very large die-stacked DRAM caches [ C ]. Proceedings of the 11 th annual Microprogramming Workshop,2011 : 123-128. 被引量:1
  • 6Rixner S,Dally W J, Kapasi U J, et al. Memory access scheduling [ C]. Proceedings of the 37th International Symposium on Comput- er Architecture ,2000:58-69. 被引量:1
  • 7Zuravleff W K, Robinson T. Controller for a synchronous DRAM that maximizes throughput by allowing memory requests and com- mands to be issued out of order[ P]. U.S. Patent 5630096,1997. 被引量:1
  • 8Eyerman S, Eeckhout L. System-level performance metrics for mul- tiprogrmn workloads[ C]. Proceedings of the 8th Annual Micropro- gramming Workshop ,2008 : 147-158. 被引量:1
  • 9Qureshi M K, Lynch D N, Mutlu O, et al. A case for MLP-aware cache replacement [ C ]. Proceedings of 33rd Annual International Symposium on Computer Architecture,2006:211-221. 被引量:1
  • 10Jiang X, Madan N, Zhao L, et al. CHOP: adaptive filter-based DRAM caching for CMP server platformsl C]. Proceedings of 16th International Symposium on High-Performance Computer Architec- ture,2010 : 165-175. 被引量:1

同被引文献10

引证文献1

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