Mitigating the Mismatch between the Coherence Protocol and Conflict Detection in Hardware Transactional Memory

被引:1
|
作者
Zhao, Lihang [1 ]
Chen, Lizhong [2 ]
Draper, Jeffrey [1 ]
机构
[1] Univ So Calif, Inst Informat Sci, Los Angeles, CA 90089 USA
[2] Univ So Calif, Ming Hsieh Dept Elect Engn, Los Angeles, CA 90007 USA
关键词
D O I
10.1109/IPDPS.2014.69
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Hardware Transactional Memory (HTM) usually piggybacks onto the cache coherence protocol to detect data access conflicts between transactions. We identify an intrinsic mismatch between the typical coherence scheme and transaction execution, which causes a sizable amount of unnecessary transaction aborts. This pathological behavior is called false aborting and increases the amount of wasted computation and on-chip communication. For the TM applications we studied, 41% of the transactional write requests incur false aborting. To combat false aborting, we propose Predictive Unicast and Notification (PUNO), a novel hardware mechanism to 1) replace the inefficient coherence multicast with a unicast scheme to prevent transactions from being disrupted unnecessarily and 2) restrain transaction polling through proactive notification. PUNO reduces transaction aborts by 61% and network traffic by 32% in workloads representative of future TM applications with a VLSI implementation area overhead of 0.41%.
引用
收藏
页数:10
相关论文
共 31 条
  • [1] Consolidated Conflict Detection for Hardware Transactional Memory
    Zhao, Lihang
    Draper, Jeffrey
    [J]. PROCEEDINGS OF THE 23RD INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT'14), 2014, : 201 - 212
  • [2] Directory-Based Conflict Detection in Hardware Transactional Memory
    Titos, Ruben
    Acacio, Manuel E.
    Garcia, Jose M.
    [J]. HIGH PERFORMANCE COMPUTING - HIPC 2008, PROCEEDINGS, 2008, 5374 : 541 - 554
  • [3] Conflict Graph Based Hardware Transactional Memory
    Zeng, Kun
    [J]. PROCEEDINGS OF 2010 3RD IEEE INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND INFORMATION TECHNOLOGY (ICCSIT 2010), VOL 5, 2010, : 496 - 501
  • [4] A Waiting Mechanism with Conflict Prediction on Hardware Transactional Memory
    Mashita, Keisuke
    Tabuchi, Maya
    Yamada, Ryohei
    Tsumura, Tomoaki
    [J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2016, E99D (12): : 2860 - 2870
  • [5] Forgive-TM: Supporting Lazy Conflict Detection In Eager Hardware Transactional Memory
    Park, Sunjae
    Hughes, Christopher J.
    Prvulovic, Milos
    [J]. 2019 28TH INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT 2019), 2019, : 192 - 204
  • [6] Speculation-Based Conflict Resolution in Hardware Transactional Memory
    Titos, Ruben
    Acacio, Manuel E.
    Garcia, Jose M.
    [J]. 2009 IEEE INTERNATIONAL SYMPOSIUM ON PARALLEL & DISTRIBUTED PROCESSING, VOLS 1-5, 2009, : 429 - 440
  • [7] Priority-Based Conflict Resolution for Hardware Transactional Memory
    Yamada, Ryohei
    Hashimoto, Koshiro
    Tsumura, Tomoaki
    [J]. 2014 SECOND INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING (CANDAR), 2014, : 433 - 439
  • [8] Using Hardware Transactional Memory for Data Race Detection
    Gupta, Shantanu
    Sultan, Florin
    Cadambi, Srihari
    Ivancic, Franjo
    Roetteler, Martin
    [J]. 2009 IEEE INTERNATIONAL SYMPOSIUM ON PARALLEL & DISTRIBUTED PROCESSING, VOLS 1-5, 2009, : 267 - +
  • [9] EcoTM: Conflict-Aware Economical Unbounded Hardware Transactional Memory
    Tomic, Sasa
    Akpinar, Ege
    Cristal, Adrian
    Unsal, Osman
    Valero, Mateo
    [J]. 2013 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE, 2013, 18 : 270 - 279
  • [10] Efficient GPU Hardware Transactional Memory through Early Conflict Resolution
    Chen, Sui
    Peng, Lu
    [J]. PROCEEDINGS OF THE 2016 IEEE INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE (HPCA-22), 2016, : 274 - 284