Hybrid approach of filtering unnecessary way accesses for set-associative caches

被引:0
|
作者
State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Science, Beijing 100190, China [1 ]
不详 [2 ]
不详 [3 ]
不详 [4 ]
不详 [5 ]
机构
[1] [1,Fan, Ling-Jun
[2] 1,Xu, Yuan-Chao
[3] Shi, Wei-Song
[4] Fan, Dong-Rui
[5] Lou, Jie
来源
Fan, L.-J. (fanlingjun@ict.ac.cn) | 1600年 / Science Press卷 / 36期
关键词
Cache architecture - Dynamic power consumption - Dynamic voltage frequency scaling - Dynamical power - Invalid filter - Law of diminishing returns - Set associative cache - Tag-2 filter;
D O I
10.3724/SP.J.1016.2013.00799
中图分类号
学科分类号
摘要
Power has been a big issue in processor design for several years. Conventional popular approaches for addressing this issue like DVFS (Dynamic Voltage Frequency Scaling) now hit the law of diminishing returns. As multi/many-core processors becoming the main stream processors, caches account for more and more CPU die area and power, this paper presents using filtering unnecessary way accesses to reduce dynamic power consumption of caches shared by instruction and data. The methods include using Invalid Filter, which could eliminate accesses to cache ways contained invalid blocks, and I/D Filter, which could eliminate accesses to cache ways contained instruction/data access type mismatch blocks, and Tag-2 Filter, which could eliminate accesses to cache ways contained tag lowest 2 bits mismatch blocks. Since the methods reducing the activities happened in cache architecture, dynamical CPU power could be significantly decreased. In the paper, we also propose combining the above methods together, which is called Invalid+I/D+Tag-2 Filter, in an attempt to achieve better power saving results. We have verified the effectiveness and complementariness of the three proposed methods through analysis and experiments. Also, our evaluations show that, we could obtain 19.6%~47.8% (which is on average 34.3%) improvement on a 64KB-4way set-associative cache and 19.6%~55.2% (which is on average 39.2%) improvement on a 128KB-8way set-associative cache comparing to Invalid+I/D Filter, and 16.1%~27.7% (which is on average 16.6%) improvement on a 64KB-4way set-associative cache and 6.9%~44.4% (which is on average 25.0%) improvement on a 128KB-8way set-associative cache comparing to Invalid+Tag-2 Filter, respectively.
引用
收藏
相关论文
共 39 条
  • [1] Way-tracking set-associative caches
    Kang, J.
    Lee, S.
    Lee, I.
    [J]. ELECTRONICS LETTERS, 2010, 46 (22) : 1497 - 1498
  • [2] Using a way cache to improve performance of set-associative caches
    Nicolaescu, Dan
    Veidenbaum, Alexander
    Nicolau, Alexandru
    [J]. HIGH-PERFORMANCE COMPUTING, 2008, 4759 : 93 - +
  • [3] Timing analysis for data caches and set-associative caches
    White, RT
    Mueller, F
    Healy, CA
    Whalley, DB
    Harmon, MG
    [J]. THIRD IEEE REAL-TIME TECHNOLOGY AND APPLICATIONS SYMPOSIUM, PROCEEDINGS, 1997, : 192 - 202
  • [4] Way Guard: A Segmented Counting Bloom Filter Approach to Reducing Energy for Set-Associative Caches
    Ghosh, Mrinmoy
    Oezer, Emre
    Ford, Simon
    Biles, Stuart
    Lee, Hsien-Hsin S.
    [J]. ISLPED 09, 2009, : 165 - 170
  • [5] An Associativity Threshold Phenomenon in Set-Associative Caches
    Bender, Michael A.
    Das, Rathish
    Farach-Colton, Martin
    Tagliavini, Guido
    [J]. PROCEEDINGS OF THE 35TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, SPAA 2023, 2023, : 117 - 127
  • [6] Generalizing timing predictions to set-associative caches
    Mueller, F
    [J]. NINTH EUROMICRO WORKSHOP ON REAL TIME SYSTEMS, PROCEEDINGS, 1997, : 64 - 71
  • [7] STACK EVALUATION OF ARBITRARY SET-ASSOCIATIVE MULTIPROCESSOR CACHES
    WU, YG
    MUNTZ, R
    [J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1995, 6 (09) : 930 - 942
  • [8] Applying decay to reduce dynamic power in set-associative caches
    Keramidas, Georgios
    Xekalakis, Polychronis
    Kaxiras, Stefanos
    [J]. HIGH PERFORMANCE EMBEDDED ARCHITECTURES AND COMPILERS, PROCEEDINGS, 2007, 4367 : 38 - +
  • [9] Recruiting Decay for Dynamic Power Reduction in Set-Associative Caches
    Keramidas, Georgios
    Xekalakis, Polychronis
    Kaxiras, Stefanos
    [J]. TRANSACTIONS ON HIGH-PERFORMANCE EMBEDDED ARCHITECTURES AND COMPILERS II, 2009, 5470 : 4 - 22
  • [10] Resilience Analysis: Tightening the CRPD bound for set-associative caches
    Altmeyer, Sebastian
    Maiza , Claire
    Reineke, Jan
    [J]. ACM SIGPLAN NOTICES, 2010, 45 (04) : 153 - 162