SWAS: Stealing Work using Approximate System-load Information

被引：0

作者：

Tzilis, Stavros ^{[1
]}

Pericas, Miquel ^{[1
]}

Trancoso, Pedro ^{[2
]}

Sourdis, Ioannis ^{[1
]}

机构：

[1] Chalmers Univ Technol, Gothenburg, Sweden

[2] Univ Cyprus, Nicosia, Cyprus

来源：

2017 46TH INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING WORKSHOPS (ICPPW) | 2017年

关键词：

D O I：

10.1109/ICPPW.2017.51

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper explores the potential of utilizing approximate system load information to enhance work stealing for dynamic load balancing in hierarchical multicore systems. Maintaining information about the load of a system has not been extensively researched since it is assumed to introduce performance overheads. We propose SWAS, a lightweight approximate scheme for retrieving and using such information, based on compact bit vector structures and lightweight update operations. This approximate information is used to enhance the effectiveness of work stealing decisions. Evaluating SWAS for a number of representative scenarios on a multi-socket multi-core platform showed that work stealing guided by approximate system load information achieves considerable performance improvements: up to 18.5% for dynamic, severely imbalanced workloads; and up to 34.4% for workloads with complex task dependencies, when compared with random work stealing.

引用

页码：309 / 318

页数：10

共 50 条

[1] Enhancement of Electrical Distribution System-Load levelling
Shrinidhi, G.
Subramanian, Durgesh M.
Rajesh, Gayathri
Kochuvila, Sreeja
[J]. PROCEEDINGS OF THE 2019 INTERNATIONAL CONFERENCE ON INTELLIGENT COMPUTING AND CONTROL SYSTEMS (ICCS), 2019, : 435 - 439
[2] Using Load Information in Work-Stealing on Distributed Systems with Non-uniform Communication Latencies
Janjic, Vladimir
Hammond, Kevin
[J]. EURO-PAR 2012 PARALLEL PROCESSING, 2012, 7484 : 155 - 166
[3] Load Balancing Using Work-stealing for Pipeline Parallelism in Emerging Applications
Navarro, Angeles
Asenjo, Rafael
Tabik, Siham
Cascaval, Calin
[J]. ICS'09: PROCEEDINGS OF THE 2009 ACM SIGARCH INTERNATIONAL CONFERENCE ON SUPERCOMPUTING, 2009, : 517 - 518
[4] LIBRA:Client Initiated Algorithm for Load Balancing Using Work Stealing Mechanism.
Deshmukh, Sudarshan S.
Deshmukh, Pooja S.
[J]. 2009 SECOND INTERNATIONAL CONFERENCE ON EMERGING TRENDS IN ENGINEERING AND TECHNOLOGY (ICETET 2009), 2009, : 67 - +
[5] Dynamic Multiple Work Stealing Strategy for Flexible Load Balancing
Adnan
Sato, Mitsuhisa
[J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2012, E95D (06): : 1565 - 1576
[6] Load Balancing Prioritized Tasks via Work-Stealing
Imam, Shams
Sarkar, Vivek
[J]. EURO-PAR 2015: PARALLEL PROCESSING, 2015, 9233 : 222 - 234
[7] PackStealLB: A scalable distributed load balancer based on work stealing and workload discretization
Freitas, Vinicius
Pilla, Laercio L.
Santana, Alexandre de L.
Castro, Marcio
Cohen, Johanne
[J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2021, 150 : 34 - 45
[8] Approximate reducts of an information system
Kuo, TF
Yajima, Y
[J]. ROUGH SETS, FUZZY SETS, DATA MINING, AND GRANULAR COMPUTING, 2003, 2639 : 291 - 294
[9] Benchmarking and Analysis of Variations of Work Stealing Scheduler on Clustered System
Kumar, Saurav
Sahu, Aryabartta
[J]. 2014 15TH INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED COMPUTING, APPLICATIONS AND TECHNOLOGIES (PDCAT 2014), 2014, : 28 - 35
[10] Scalable Work-Stealing Load-Balancer for HPC Distributed Memory Systems
Fontenaille, Clement
Petit, Eric
Castro, Pablo de Oliveira
Uemura, Seijilo
Sohier, Devan
Lesnicki, Piotr
Lartigue, Ghislain
Moureau, Vincent
[J]. EURO-PAR 2018: PARALLEL PROCESSING WORKSHOPS, 2019, 11339 : 146 - 158

← 1 2 3 4 5 →