Journal of the Institute of Electronics Engineers of Korea CI (전자공학회논문지CI)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Filter Cache Predictor Using Mode Selection Bit

모드 선택 비트를 사용한 필터 캐시 예측기

  • Kwak, Jong-Wook (Department of Computer Engineering, Yeungnam University)
  • 곽종욱 (영남대학교 컴퓨터공학과)
  • Published : 2009.09.25
Download PDF
⟨ Previous Next ⟩

Abstract

Filter cache has been introduced as one solution of reducing cache power consumption. More than 50% of the power reduction results from the filter cache, whereas more than 20% of the performance is compromised. To minimize the performance degradation of the filter cache, the predictive filter cache has been proposed. In this paper, we review the previous filter cache predictors and analyze the problems of the solutions. As a result, we found main problems that cause prediction misses in previous filter cache schemes and, to resolve the problems, this paper proposes a new prediction policy. In our scheme, some reference bit entries, called MSBs, are inserted into filter cache and BTB, to adaptively control the filter cache access. In simulation parts, we use a modified SimpleScalar simulator with MiBench benchmark programs to verify the proposed filter cache. The simulation result shows in average 5% performance improvement, compared to previous ones.

캐시 에너지의 소비 전력을 줄이기 위해 필터 캐시가 제안되었다. 이와 같은 필터 캐시의 사용으로 인해 50% 이상의 전력 사용 감소 효과를 가져왔으나, 상대적으로 시스템 성능은 평균 20% 가량 감소되었다. 필터 캐시의 사용으로 인한 이 같은 성능 감소를 최소화하기 위해서, 여러 가지 형태의 필터 캐시 예측가 제안 되었다. 본 논문에서는 기존에 제안된 주요 필터 캐시 예측 모델들을 소개하며, 각각의 방식에 있어서의 핵심 특징 및 해당 방식의 문제점을 분석한다. 분석 결과, 필터 캐시의 참조 실패를 야기하는 기존 방식의 중요한 문제점을 확인하였으며, 이를 바탕으로 본 논문에서는 개선된 형태의 새로운 필터 캐시 예측기 모델을 제안한다. 제안된 방식은 MSB라 불리는 참조 비트를 고안하여 이를 기존의 필터캐시와 BTB에 새롭게 활용한다. 본 논문에서 제안된 방식의 성능을 검증하기 위해 SimpleScalar 시뮬레이터와 MiBench 응용 프로그램을 활용하여 모의실험을 수행하였다. 실험 결과 제안된 방식은 기존 방식 대비, 필터 캐시 예측 실패율, 필터 캐시 활용률 및 전력 소모량 시간 지연 등 모든 면에서 평균 5%의 성능 향상을 가져 왔다.

Keywords

References

  1. D. Patterson, and J. Hennessy, 'Computer architecture: a quantitative approach' 4th edition, Morgan Kaufman, 2007
  2. J. Montanaro et al. 'A 160-MHz, 32-b, 0.5-W CMOS RISC microprocessor', IEEE Journal of Solid-State Circuits, 32(11):1703-14, 1996
  3. C.-L. Su and A. Despain. 'Cache design tradeoffs for power and performance optimization: A case study', In Proceedings of International Symposium on Low Power Design, April 1995 https://doi.org/10.1145/224081.224093
  4. W. Tang, A. Kejariwal, A. Veidenbaum and A. Nicolau, 'A Predictive Decode Filter Cache for Reducing Power Consumption in Embedded Processors', ACM Transactions on Design Automation of Electronic Systems, Vol. 12, No. 2, April 2007 https://doi.org/10.1145/1230800.1230806
  5. P. P. Chang, S. A. Mahlke, W. Y. Chen, N. J.Warter, and W.-m. W. Hwu, 'IMPACT: An Architectural Framework for Multiple- Instruction-Issue Processors', Proc. of ISCA, 1991
  6. C. Lee, M. Potkonjak, and W. H. Mangione- Smith, 'MediaBench: A Tool for Evaluating Multimedia and Communications Systems', Proc. of Micro 30, 1997
  7. Anderson, T., and Agarwala, S.: 'Effective hardware-based two-way loop cache for high- performance low-power processors', Proc. Int. Conf. on Computer Design, pp. 403–407, 2000
  8. Tang, W., Gupta, R., and Nicolau, A.: 'Design of a predictive filter cache for energy savings in high-performance processor architectures'. Proc. Int. Conf. on Computer Design, pp. 68–73, 2001
  9. Yeh, T.Y., and Patt, Y.N.: 'Alternative implementation of two-level adaptive branch prediction', Proc. 19th Int. Symp. on Computer Architecture, pp. 124–124, 1992
  10. K. Vivekanandarajah, T. Srikanthan, S. Bhattacharyya, 'Energy-delay effficient filter cache hierarchy using pattern prediction scheme', IEE Proceedings - Computers and Digital Techniques, Vol. 151, Issue 2, March 2004
  11. Ernst, D., Austin, T.M., Mudge, T., and Brown, R.B. 'MiBench: a free commercially representative embedded benchmark suite', Proc. 4th IEEE Int. Workshop on Workload characterization, pp. 3–14, Dec. 2001
  12. Chia-Lin Yang, Chien-Hao Lee, 'HotSpot cache: joint temporal and spatial locality exploitation for i-cache energy reduction', ISLPED pp. 114-119, 2004
  13. P.-Y. Chang, M. Evers, and Y. Patt. 'Improving Branch Prediction Accuracy by Reducing Pattern History Table Interference', Proc. Int. Conf. on Parallel Architectures and Compilation Techniques, Oct. 1996
  14. D. Burger, A. Kagi, and M. Hrishikesh. 'Memory hierarchy extensions to SimpleScalar 3.0', Technical Report TR99-25, Department of Computer Science, University of Texas at Austin, April 1999
  15. David Tarjan, Shyamkumar Thoziyoor, Norman P. Jouppi, 'CACTI 4.0', HP Laboratories Palo Alto HPL-2006-86, June 2, 2006
  16. B. Case. SPEC2000 Retires SPEC92, The Microprocessor Report, vol. 9, 1995
  17. EDN Embedded Microprocessor Benchmark Consortium, http://www.eembc.org
  18. K. Vivekanandarajah et al., 'Incorporating pattern prediction technique for energy efficient filter cache design', The 3rd IEEE International Workshop on SoC for Real-Time Applications, Vol. 30, pp. 44-47, 2003