Energy-efficient MESI cache coherence with pro-active snoop filtering for multicore microprocessors

ABSTRACT

We present a snoop filtering mechanism for multicore microprocessors that implement coherent caches using the MESI protocol. The relatively small filter structure at each core maintains coarse-grain sharing information about regions within a page to filter out snoops. On broadcast, the sharing status of all regions within the page is collected proactively and up to 90% of unnecessary snoops are eliminated. The energy savings resulting from snoop filtering in our scheme average about 30% across the benchmarks studied for both a quad core design in 65 nm and 8-core design in 45 nm CMOS.

REFERENCES

• 1: Anderson, J. and Jayasimha, D. N., "Snoop Filter Line Replacement of Back Invalidates in Multi-Node Architectures", US Patent No. 6,598,123 B1, July 22, 2003.
• 2: Fayé Briggs , Suresh Chittor , Kai Cheng, Micro-architecture techniques in the intel® E8870 scalable memory controller, Proceedings of the 3rd workshop on Memory performance issues: in conjunction with the 31st international symposium on computer architecture, p.30-36, June 20-20, 2004, Munich, Germany [doi>10.1145/1054943.1054948]
• 3: Jason F. Cantin , Mikko H. Lipasti , James E. Smith, Improving Multiprocessor Performance with Coarse-Grain Coherence Tracking, Proceedings of the 32nd annual international symposium on Computer Architecture, p.246-257, June 04-08, 2005
• 4: Cantin, J. F. et al, "Coarse-Grain Coherence Tracking: RegionSCout and Region Coherence Arrays", Computer, June 2006.
• 5: Chintamani, S. and Iyer, R., "Design and Evaluation of Snoop Filters for Web Servers", in Proc. SPECTS '04, 2004.
• 6: M. Ghoneima , Y. Ismail , M. Khellah , J. Tschanz , V. De, Serial-link bus: a low-power on-chip bus architecture, Proceedings of the 2005 IEEE/ACM International conference on Computer-aided design, p.541-546, November 06-10, 2005, San Jose, CA
• 7: Magnus Ekman , Per Stenström , Fredrik Dahlgren, TLB and snoop energy-reduction using virtual caches in low-power chip-multiprocessors, Proceedings of the 2002 international symposium on Low power electronics and design, August 12-14, 2002, Monterey, California, USA [doi>10.1145/566408.566471]
• 8: Milo M. K. Martin , Daniel J. Sorin , Bradford M. Beckmann , Michael R. Marty , Min Xu , Alaa R. Alameldeen , Kevin E. Moore , Mark D. Hill , David A. Wood, Multifacet's general execution-driven multiprocessor simulator (GEMS) toolset, ACM SIGARCH Computer Architecture News, v.33 n.4, November 2005 [doi>10.1145/1105734.1105747]
• 9: Kanter, D., "x86 Servers Brace for a Hurricane", available from http://www.realwordtech.com, April 2004.
• 10: Mamidipika, M.., "eCACTI: An Enhanced Power Estimation Model for On-Chip Caches", CECS TR , UC-Irvine, 2004.
• 11: Andreas Moshovos , Gokhan Memik , Alok Choudhary , Babak Falsafi, JETTY: Filtering Snoops for Reduced Energy Consumption in SMP Servers, Proceedings of the 7th International Symposium on High-Performance Computer Architecture, p.85, January 20-24, 2001
• 12: Andreas Moshovos, RegionScout: Exploiting Coarse Grain Sharing in Snoop-Based Coherence, Proceedings of the 32nd annual international symposium on Computer Architecture, p.234-245, June 04-08, 2005
• 13: Saldanha, C. andLipasti, M., "Power-Efficient Cache Coherence", in Proc. Workshop on Memory Performance Issues, ISCA 2001.
• 14: Karin Strauss , Xiaowei Shen , Josep Torrellas, Flexible Snooping: Adaptive Forwarding and Filtering of Snoops in Embedded-Ring Multiprocessors, Proceedings of the 33rd annual international symposium on Computer Architecture, p.327-338, June 17-21, 2006
• 15: Jason Zebchuk , Elham Safi , Andreas Moshovos, A Framework for Coarse-Grain Optimizations in the On-Chip Memory Hierarchy, Proceedings of the 40th Annual IEEE/ACM International Symposium on Microarchitecture, p.314-327, December 01-05, 2007 [doi>10.1109/MICRO.2007.5]

INDEX TERMS

Primary Classification:
  B. Hardware
  B.3 MEMORY STRUCTURES
      B.3.2 Design Styles
          Subjects: Cache memories

General Terms:
Design

Keywords:
cache coherence, multicore microprocessor