Abstract
This paper proposes an architecture for low-power direct-mapped instruction caches, called “history-based tag-comparison (HBTC) cache”. The HBTC cache attempts to detect and omit unnecessary tag checks at run time. Execution footprints are recorded in an extended BTB (Branch Target Buffer), and are used to know the cache residence of target instructions before starting cache access. In our simulation, it is observed that our approach can reduce the total count of tag checks by 90 %, resulting in 15 % of cache-energy reduction, with less than 0.5 % performance degradation.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Bahar, I., Albera, G., and Manne, S.: Power and Performance Tradeoffs using Various Caching Strategies. Proc. of the 1998 International Symposium on Low Power Electronics and Design, pp. 64–69, Aug. 1998.
Bellas, N., Hajj, I., and Polychronopoulos, C.: Using dynamic cache management techniques to reduce energy in a high-performance processor. Proc. of the 1999 International Symposium on Low Power Electronics and Design, pp. 64–69, Aug. 1999.
Bellas, N., Hajj, I., Polychronopoulos. C., and Stamoulis, G.: Energy and Performance Improvements in Microprocessor Design using a Loop Cache. Proc. of the 1999 International Conference on Computer Design: VLSI in Computers & Processors, pp. 378–383, Oct. 1999.
Inoue, K. and Murakami, K.: A Low-Power Instruction Cache Architecture Exploiting Program Execution Footprints. International Symposium on High-Performance Computer Architecture, Work-in-progress session (included in the CD proceedings), Feb. 2001.
Ishihara, T. and Yasuura, H.: A Power Reduction Technique with Object Code Merging for Application Specific Embedded Processors. Proc. of the Design, Automation and Test in Europe Conference, pp617–623, Mar. 2000.
Kamble, M. and Ghose, K.: Analytical Energy Dissipation Models For Low Power Caches. Proc. of the 1997 International Symposium on Low Power Electronics and Design, pp. 143–148, Aug. 1997.
Kamble, M. and Ghose, K.: Energy-Efficiency of VLSI Caches: A Comparative Study. Proc. of the 10th International Conference on VLSI Design, pp. 261–267, Jan. 1997.
Kin, J., Gupta, M., and Mngione-Smith, W.: The Filter Cache: An Energy Efficient Memory Structure. Proc. of the 30th Annual International Symposium on Microarchitecture, pp. 184–193, Dec. 1997.
Ma, A., Zhang, M., and Asanović, K.: Way Memorization to Reduce Fetch Energy in Instruction Caches. ISCA Workshop on Complexity Effective Design, July 2001.
Panda, R., Dutt, N., and Nicolau, A.: Efficient Utilization of Scratch-Pad Memory in Embedded Processor Applications. Proc. of European Design & Test Conference, Mar. 1997.
Panwar, R. and Rennels, D.: Reducing the frequency of tag compares for low power I-cache design. Proc. of the 1995 International Symposium on Low Power Electronics and Design, Aug. 1995.
Wilton, S. and Jouppi, N.: An Enhanced Access and Cycle Time Model for On-Chip Caches. WRL Research Report 93/5, July 1994.
Witchel, E., Larsen, S., Ananian, C., and Asanović, K.: Direct Addressed Caches for Reduced Power Consumption. Proc. of the 34th International Symposium on Microarchitecture, Dec. 2001.
MediaBench, URL: http://www.cs.ucla.edu/~leec/mediabench/.
“SimpleScalar Simulation Tools for Microprocessor and System Evaluation,” URL:http://www.simplescalar.org/.
SPEC (Standard Performance Evaluation Corporation), URL: http://www.specbench.org/osg/cpu95.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Inoue, K., Moshnyaga, V., Murakami, K. (2003). Dynamic Tag-Check Omission: A Low Power Instruction Cache Architecture Exploiting Execution Footprints. In: Falsafi, B., Vijaykumar, T.N. (eds) Power-Aware Computer Systems. PACS 2002. Lecture Notes in Computer Science, vol 2325. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36612-1_2
Download citation
DOI: https://doi.org/10.1007/3-540-36612-1_2
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-01028-9
Online ISBN: 978-3-540-36612-6
eBook Packages: Springer Book Archive