Stride-directed prefetching for secondary caches

被引:8
|
作者
Kim, S
Veidenbaum, AV
机构
来源
PROCEEDINGS OF THE 1997 INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING | 1997年
关键词
D O I
10.1109/ICPP.1997.622661
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
This paper studies hardware prefetcheng for second-level (L2) caches. Previous work on prefetching has been extensive but largely directed at primary caches. In some cases only L2 prefetching is possible or is more appropriate. By studying L2 prefetching characteristics we show that existing stride-directed methods [1, 8] for L1 caches do not work as well in L2 caches. We propose a new stride-detection mechanism for L2 prefetching and combine it with stream buffers used in [16]. Our evaluation shows that this new prefetching scheme is more effective than stream buffer prefetching particularly for applications with long-stride accesses. Finally, we evaluate an L2 cache prefetching organization which combines a small L2 cache with our stride-directed prefetching scheme. Our results show that this system performs significantly better than stream buffer prefetching or a larger non-prefetching L2 cache without suffering from a significant increase in the memory traffic.
引用
收藏
页码:314 / 321
页数:8
相关论文
共 50 条
  • [41] ARE SECONDARY CACHES FOR THE 486 WORTH THE MONEY
    WILSON, D
    COMPUTER DESIGN, 1991, 30 (08): : 28 - +
  • [42] Efficient discovery of regular stride patterns in irregular programs and its use in compiler prefetching
    Wu, YF
    ACM SIGPLAN NOTICES, 2002, 37 (05) : 210 - 221
  • [43] Evaluation of hardware-based stride and sequential prefetching in shared-memory multiprocessors
    Dahlgren, F
    Stenstrom, P
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1996, 7 (04) : 385 - 398
  • [44] A decoupled predictor-directed stream prefetching architecture
    Sair, S
    Sherwood, T
    Calder, B
    IEEE TRANSACTIONS ON COMPUTERS, 2003, 52 (03) : 260 - 276
  • [45] Feedback directed prefetching optimization for linked data structure
    Jiangnan Institute of Computing Technology, Wuxi 214083, China
    Ruan Jian Xue Bao, 2009, SUPPL. 1 (34-39):
  • [46] A stateless, content-directed data prefetching mechanism
    Cooksey, R
    Jourdan, S
    Grunwald, D
    ACM SIGPLAN NOTICES, 2002, 37 (10) : 279 - 290
  • [47] EFFECTIVENESS OF COMPILER-DIRECTED PREFETCHING ON DATA MINING BENCHMARKS
    Natarajan, Ragavendra
    Mekkat, Vineeth
    Hsu, Wei-Chung
    Zhai, Antonia
    JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS, 2012, 21 (02)
  • [48] Cost-effective compiler directed memory prefetching and bypassing
    Ortega, D
    Ayguadé, E
    Baer, JL
    Valero, M
    2002 INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES, PROCEEDINGS, 2002, : 189 - 198
  • [49] A Compiler-Directed Data Prefetching Scheme for Chip Multiprocessors
    Son, Seung Woo
    Kandemir, Mahmut
    Karakoy, Mustafa
    Chakrabarti, Dhruva
    ACM SIGPLAN NOTICES, 2009, 44 (04) : 209 - 218
  • [50] UDP: Utility-Driven Fetch Directed Instruction Prefetching
    Oh, Surim
    Xu, Mingsheng
    Khan, Tanvir Ahmed
    Kasikci, Baris
    Litz, Heiner
    2024 ACM/IEEE 51ST ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE, ISCA 2024, 2024, : 1188 - 1201