Complexity-effective contention management with dynamic backoff for transactional memory systems

Seung Hun Kim, Dongmin Choi, Won Woo Ro, Jean Luc Gaudiot

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Reducing memory access conflicts is a crucial part of the design of Transactional Memory (TM) systems since the number of running threads increases and long latency transactions gradually appear: without an efficient contention management, there will be repeated aborts and wasteful rollback operations. In this paper, we present a dynamic backoff control algorithm developed for complexity-effective and distributed contention management in Hardware Transactional Memory (HTM) systems. Our approach aims at controlling the restarting intervals of aborted transactions, and can be easily applied to the various TM systems. To this end, we have profiled the applications of the STAMP benchmark suite and have identified those 'problem' transactions which repeatedly cause aborts in the applications with the attendant high contention rate. The proposed algorithm alleviates the impact of these repeated aborts by dynamically adjusting the initial exponent value of the traditional backoff approach. In addition, the proposed scheme decreases the number of wasted cycles down to 82% on average compared to the baseline TM system. Our design has been integrated in LogTM-SE where we observed an average performance improvement of 18%.

Original languageEnglish
Article number6475127
Pages (from-to)1696-1708
Number of pages13
JournalIEEE Transactions on Computers
Volume63
Issue number7
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Transactional Memory
Contention
Computer systems
Transactions
Data storage equipment
Dynamic Algorithms
Dynamic Control
Thread
Control Algorithm
Latency
Baseline
Exponent
Computer hardware
Hardware
Benchmark
Cycle
Decrease
Interval
Design

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

Kim, Seung Hun ; Choi, Dongmin ; Ro, Won Woo ; Gaudiot, Jean Luc. / Complexity-effective contention management with dynamic backoff for transactional memory systems. In: IEEE Transactions on Computers. 2014 ; Vol. 63, No. 7. pp. 1696-1708.
@article{8ef32e79827842b3b64d9c22c5265a8c,
title = "Complexity-effective contention management with dynamic backoff for transactional memory systems",
abstract = "Reducing memory access conflicts is a crucial part of the design of Transactional Memory (TM) systems since the number of running threads increases and long latency transactions gradually appear: without an efficient contention management, there will be repeated aborts and wasteful rollback operations. In this paper, we present a dynamic backoff control algorithm developed for complexity-effective and distributed contention management in Hardware Transactional Memory (HTM) systems. Our approach aims at controlling the restarting intervals of aborted transactions, and can be easily applied to the various TM systems. To this end, we have profiled the applications of the STAMP benchmark suite and have identified those 'problem' transactions which repeatedly cause aborts in the applications with the attendant high contention rate. The proposed algorithm alleviates the impact of these repeated aborts by dynamically adjusting the initial exponent value of the traditional backoff approach. In addition, the proposed scheme decreases the number of wasted cycles down to 82{\%} on average compared to the baseline TM system. Our design has been integrated in LogTM-SE where we observed an average performance improvement of 18{\%}.",
author = "Kim, {Seung Hun} and Dongmin Choi and Ro, {Won Woo} and Gaudiot, {Jean Luc}",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/TC.2013.49",
language = "English",
volume = "63",
pages = "1696--1708",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "7",

}

Complexity-effective contention management with dynamic backoff for transactional memory systems. / Kim, Seung Hun; Choi, Dongmin; Ro, Won Woo; Gaudiot, Jean Luc.

In: IEEE Transactions on Computers, Vol. 63, No. 7, 6475127, 01.01.2014, p. 1696-1708.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Complexity-effective contention management with dynamic backoff for transactional memory systems

AU - Kim, Seung Hun

AU - Choi, Dongmin

AU - Ro, Won Woo

AU - Gaudiot, Jean Luc

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Reducing memory access conflicts is a crucial part of the design of Transactional Memory (TM) systems since the number of running threads increases and long latency transactions gradually appear: without an efficient contention management, there will be repeated aborts and wasteful rollback operations. In this paper, we present a dynamic backoff control algorithm developed for complexity-effective and distributed contention management in Hardware Transactional Memory (HTM) systems. Our approach aims at controlling the restarting intervals of aborted transactions, and can be easily applied to the various TM systems. To this end, we have profiled the applications of the STAMP benchmark suite and have identified those 'problem' transactions which repeatedly cause aborts in the applications with the attendant high contention rate. The proposed algorithm alleviates the impact of these repeated aborts by dynamically adjusting the initial exponent value of the traditional backoff approach. In addition, the proposed scheme decreases the number of wasted cycles down to 82% on average compared to the baseline TM system. Our design has been integrated in LogTM-SE where we observed an average performance improvement of 18%.

AB - Reducing memory access conflicts is a crucial part of the design of Transactional Memory (TM) systems since the number of running threads increases and long latency transactions gradually appear: without an efficient contention management, there will be repeated aborts and wasteful rollback operations. In this paper, we present a dynamic backoff control algorithm developed for complexity-effective and distributed contention management in Hardware Transactional Memory (HTM) systems. Our approach aims at controlling the restarting intervals of aborted transactions, and can be easily applied to the various TM systems. To this end, we have profiled the applications of the STAMP benchmark suite and have identified those 'problem' transactions which repeatedly cause aborts in the applications with the attendant high contention rate. The proposed algorithm alleviates the impact of these repeated aborts by dynamically adjusting the initial exponent value of the traditional backoff approach. In addition, the proposed scheme decreases the number of wasted cycles down to 82% on average compared to the baseline TM system. Our design has been integrated in LogTM-SE where we observed an average performance improvement of 18%.

UR - http://www.scopus.com/inward/record.url?scp=84903648246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84903648246&partnerID=8YFLogxK

U2 - 10.1109/TC.2013.49

DO - 10.1109/TC.2013.49

M3 - Article

VL - 63

SP - 1696

EP - 1708

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 7

M1 - 6475127

ER -