PleaseTM: Enabling Transaction Conflict Management in Requester-wins Hardware Transactional Memory

被引:0
|
作者
Park, Sunjae [1 ]
Prvulovic, Milos [1 ]
Hughes, Christopher J. [2 ]
机构
[1] Georgia Inst Technol, Atlanta, GA 30332 USA
[2] Intel, Hillsboro, OR USA
基金
美国国家科学基金会;
关键词
D O I
暂无
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
With recent commercial offerings, hardware transactional memory (HTM) has finally become an important tool in writing multithreaded applications. However, current offerings are commonly implemented in a way that keep the coherence protocol unmodified. Data conflicts are recognized by coherence messages sent by the requester to sharers of the cache block (e.g., a write to a speculatively read line), who are then aborted. This tends to abort transactions that have done more work, leading to suboptimal performance. Even worse, this can lead to live-lock situations where transactions repeatedly abort each other. In this paper, we present PleaseTM, a mechanism that allows more freedom in deciding which transaction to abort, while leaving the coherence protocol design unchanged. In PleaseTM, transactions insert plea bits into their responses to coherence requests as a simple payload, and use these bits to inform conflict management decisions. Coherence permission changes are then achieved with normal coherence requests. Our experiments show that this additional freedom can provide on average 43% speedup, with a maximum of 7-fold speedup, on STAMP benchmarks running at 32 threads compared to requester-wins HTM.
引用
下载
收藏
页码:285 / 296
页数:12
相关论文
共 32 条
  • [11] Priority-Based Conflict Resolution for Hardware Transactional Memory
    Yamada, Ryohei
    Hashimoto, Koshiro
    Tsumura, Tomoaki
    2014 SECOND INTERNATIONAL SYMPOSIUM ON COMPUTING AND NETWORKING (CANDAR), 2014, : 433 - 439
  • [12] Multiversioned Page Overlays: Enabling Faster Serializable Hardware Transactional Memory
    Wang, Ziqi
    Kozuch, Michael A.
    Mowry, Todd C.
    Seshadri, Vivek
    2019 28TH INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT 2019), 2019, : 394 - 407
  • [13] A Comprehensive Scheme for Contention Management in Hardware Transactional Memory
    Wang, Xiaoqun
    Ji, Zhenzhou
    Fu, Chen
    Hu, Mingzeng
    INFORMATION AND AUTOMATION, 2011, 86 : 397 - 403
  • [14] Implementation of Hybrid Version Management in Hardware Transactional Memory
    Zhao, Lihang
    Draper, Jeff
    2013 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS), 2013, : 777 - 780
  • [15] EcoTM: Conflict-Aware Economical Unbounded Hardware Transactional Memory
    Tomic, Sasa
    Akpinar, Ege
    Cristal, Adrian
    Unsal, Osman
    Valero, Mateo
    2013 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE, 2013, 18 : 270 - 279
  • [16] Efficient GPU Hardware Transactional Memory through Early Conflict Resolution
    Chen, Sui
    Peng, Lu
    PROCEEDINGS OF THE 2016 IEEE INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE (HPCA-22), 2016, : 274 - 284
  • [17] Efficient Eager Management of Conflicts for Scalable Hardware Transactional Memory
    Titos-Gil, Ruben
    Acacio, Manuel E.
    Garcia, Jose M.
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2013, 24 (01) : 59 - 71
  • [18] Efficient Management of Speculative Data in Hardware Transactional Memory Systems
    Waliullah, M. M.
    Stenstrom, Per
    2008 INTERNATIONAL CONFERENCE ON EMBEDDED COMPUTER SYSTEMS: ARCHITECTURES, MODELING AND SIMULATION, PROCEEDINGS, 2008, : 158 - 164
  • [19] GCMS: A Global Contention Management Scheme in Hardware Transactional Memory
    Wang, Xiaoqun
    Ji, Zhenzhou
    Fu, Chen
    Hu, Mingzeng
    IEEE COMPUTER ARCHITECTURE LETTERS, 2011, 10 (01) : 24 - 27
  • [20] Yet Another Waiting Mechanism Based on Conflict Prediction for Hardware Transactional Memory
    20161902341199
    (1) Nagoya Institute of Technology, Gokiso, Showa, Nagoya, Japan, 1600, (Institute of Electrical and Electronics Engineers Inc., United States):