A case for multi-level main memory

ABSTRACT

Current trends suggest that the number of memory chips per processor chip will increase at least a factor of ten in seven years. This will make DRAM cost, the space and the power it consumes a serious problem. The main question raised in this research is how cost, size, and power consumption can be reduced by transforming traditional flat main-memory systems into a multi-level hierarchy. We make the case for a multi-level main memory hierarchy by proposing and evaluating the performance of an implementation that enables aggressive use of memory compression, sharing of memory resources among computers, and dynamic power management of unused regions of memory. This paper presents the key design strategies to make this happen. We evaluate our implementation using complete runs of applications from the Spec 2K suite, SpecJBB, and SAP --- typical desktop and server applications. We show that only 30% of the entire memory resources typically needed must be accessed at DRAM speed whereas the rest can be accessed at a speed that is a magnitude slower. The resulting performance overhead is shown to be only 1.2% on average.

REFERENCES

• 1: Bulent Abali , Hubertus Franke , Xiaowei Shen , Dan E. Poff , T. Basil Smith, Performance of Hardware Compressed Main Memory, Proceedings of the 7th International Symposium on High-Performance Computer Architecture, p.73, January 20-24, 2001
• 2: Erik G. Hallnor , Steven K. Reinhardt, A fully associative software-managed cache design, Proceedings of the 27th annual international symposium on Computer architecture, p.107-116, June 2000, Vancouver, British Columbia, Canada
• 3: John L. Hennessy , David A. Patterson, Computer Architecture: A Quantitative Approach, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2003
• 4: IBM. POWER4 System Microarchitecture. Technical White Paper, October 2001.
• 5: Kalla R, Sinharoy B and Tendler J. Simultaneous Multi-threading Implementation in POWER5 -- IBM's Next Generation POWER Microprocessor. Hot Chips 15, August 2003.
• 6: Kjelso M, Gooch M and Jones S. Performance evaluation of computer architectures with main memory data compression. Journal of Systems Architecture. vol 45, pp. 571--590, 1999.
• 7: Alvin R. Lebeck , Xiaobo Fan , Heng Zeng , Carla Ellis, Power aware page allocation, Proceedings of the ninth international conference on Architectural support for programming languages and operating systems, p.105-116, November 2000, Cambridge, Massachusetts, United States
• 8: Mattson R, Gecsei J, Slutz D and Traiger I. Evaluation techniques for storage hierarchies. IBM Systems Journal, vol. 9, no. 2, pp. 78--117, 1970.
• 9: Jim Mauro , Richard McDougall, Solaris internals: core kernel architecture, Sun Microsystems, Inc., Mountain View, CA, 2000
• 10: José Luis Núñez , Simon Jones, Gbit/s lossless data compression hardware, IEEE Transactions on Very Large Scale Integration (VLSI) Systems, v.11 n.3, p.499-510, June 2003 [doi>10.1109/TVLSI.2003.812288]
• 11: Personal communications with Núnes J, author of {10}.
• 12: Ashley Saulsbury , Su-Jaen Huang , Fredrik Dahlgren, Efficient management of memory hierarchies in embedded DRAM systems, Proceedings of the 13th international conference on Supercomputing, p.464-473, June 20-25, 1999, Rhodes, Greece [doi>10.1145/305138.305236]
• 13: Sun Microsystems. Introduction to Throughput Computing. Technical White Paper, February 2003.
• 14: Sun Microsystems. UltraSPARC® IV Processor Architecture Overview. Technical White Paper, February 2004.
• 15: Takumi Maruyama. SPARC64 VI: Fujitsu's Next Generation Processor. Microprocessor Forum, October 2003.
• 16: P. Wilson, S. Kaplan, and Y. Smaragdakis. The Case for Compressed Caching in Virtual Memory Systems. In USENIX Ann. Technical Conference. June 1999.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.0 GENERAL

General Terms:
Design, Performance

Keywords:
memory-systems