Performance Evaluation of Thread-Level Speculation in Off-the-Shelf Hardware Transactional Memories

被引:5
|
作者
Salamanca, Juan [1 ]
Amaral, Jose Nelson [2 ]
Araujo, Guido [1 ]
机构
[1] Univ Estadual Campinas, Inst Comp, Campinas, SP, Brazil
[2] Univ Alberta, Comp Sci Dept, Edmonton, AB, Canada
来源
EURO-PAR 2017: PARALLEL PROCESSING | 2017年 / 10417卷
基金
巴西圣保罗研究基金会; 加拿大自然科学与工程研究理事会;
关键词
Thread-Level Speculation; Transactional memory;
D O I
10.1007/978-3-319-64203-1_44
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Thread-Level Speculation (TLS) is a hardware/software technique that enables the execution of multiple loop iterations in parallel, even in the presence of some loop-carried dependences. TLS requires hardware mechanisms to support conflict detection, speculative storage, in-order commit of transactions, and transaction roll-back. There is no off-the-shelf processor that provides direct support for TLS. Speculative execution is supported, however, in the form of Hardware Transactional Memory (HTM)-available in recent processors such as the Intel Core and the IBM POWER8. Earlier work has demonstrated that, in the absence of specific TLS support in commodity processors, HTM support can be used to implement TLS. This paper presents a careful evaluation of the implementation of TLS on the HTM extensions available in such machines. This evaluation provides evidence to support several important claims about the performance of TLS over HTM in the Intel Core and the IBM POWER8 architectures. Experimental results reveal that by implementing TLS on top of HTM, speed-ups of up to 3.8x can be obtained for some loops.
引用
收藏
页码:607 / 621
页数:15
相关论文
共 50 条
  • [1] Thread-Level Speculation on Off-the-Shelf Hardware Transactional Memory
    Odaira, Rei
    Nakaike, Takuya
    2014 IEEE INTERNATIONAL SYMPOSIUM ON WORKLOAD CHARACTERIZATION (IISWC), 2014, : 212 - 221
  • [2] Evaluating and Improving Thread-Level Speculation in Hardware Transactional Memories
    Salamanca, Juan
    Amaral, Jose Nelson
    Araujo, Guido
    2016 IEEE 30TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS 2016), 2016, : 586 - 595
  • [3] Hardware Thread-Level Speculation Performance Analysis
    Wang, Ying-Chieh
    Chung, I-Hsin
    Lee, Che-Rung
    Perrone, Michael
    Chung, Yeh-Ching
    2015 IEEE 17TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING AND COMMUNICATIONS, 2015 IEEE 7TH INTERNATIONAL SYMPOSIUM ON CYBERSPACE SAFETY AND SECURITY, AND 2015 IEEE 12TH INTERNATIONAL CONFERENCE ON EMBEDDED SOFTWARE AND SYSTEMS (ICESS), 2015, : 900 - 903
  • [4] Performance Modeling for Hardware Thread-Level Speculation
    Wang, Ying-Chieh
    Lee, Che-Rung
    Chung, Yeh-Ching
    Chung, I-Hsin
    Perrone, Michael
    PROCEEDINGS OF 2014 IEEE INTERNATIONAL PARALLEL & DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW), 2014, : 1458 - 1465
  • [5] Thread-Level Speculation Based on Transactional Memory
    Li, Xiang
    Zhang, Jing
    INFORMATION-AN INTERNATIONAL INTERDISCIPLINARY JOURNAL, 2012, 15 (04): : 1745 - 1755
  • [6] Using Hardware-Transactional-Memory Support to Implement Thread-Level Speculation
    Salamanca, Juan
    Amaral, Jose Nelson
    Araujo, Guido
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2018, 29 (02) : 466 - 480
  • [7] Performance Comparison of Speculative Taskloop and OpenMP-for-Loop Thread-Level Speculation on Hardware Transactional Memory
    Salamanca, Juan
    2022 21ST INTERNATIONAL SYMPOSIUM ON PARALLEL AND DISTRIBUTED COMPUTING (ISPDC 2022), 2022, : 83 - 90
  • [8] A scalable approach to Thread-Level Speculation
    Steffan, JG
    Colohan, CB
    Zhai, A
    Mowry, TC
    PROCEEDING OF THE 27TH INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE, 2000, : 1 - 12
  • [9] Exclusive Squashing for Thread-Level Speculation
    Garcia-Yagueez, Alvaro
    Llanos, Diego R.
    Gonzalez-Escribano, Arturo
    HPDC 11: PROCEEDINGS OF THE 20TH INTERNATIONAL SYMPOSIUM ON HIGH PERFORMANCE DISTRIBUTED COMPUTING, 2011, : 275 - 276
  • [10] Loop recreation for thread-level speculation
    Gao, Lin
    Li, Lian
    Xue, Jingling
    Ngai, Tin-Fook
    2007 INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS, VOLS 1 AND 2, 2007, : 337 - +