A multilevel partitioning approach for efficient tasks allocation in heterogeneous distributed systems

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

This work addresses the problem of allocating parallel application tasks to heterogeneous distributed computing resources, such as multiclusters or Grid environments. The proposed allocation scheme is based on a multilevel graph partitioning and mapping approach. The objective is to find an efficient allocation that minimizes the application completion time, subject to the specified constraints pertinent to the application and system environment. The allocation scheme consists of three phases; the clustering phase, the initial mapping phase and the refinement and remapping phase. The scheme introduces an efficient heuristic in the clustering phase for contracting (coarsening) large size application graphs to the number of processors, called the VHEM method. An initial mapping technique based on a tabu-search approach has been introduced as a basis for the process of refinement and remapping phase. The simulation study shows that the VHEM coarsening heuristic can achieve optimal or near-optimal communication, compared to the HEM method, when the ratio of the number of tasks to the number of processors exceeds a threshold value. The simulation study shows that those optimal or near-optimal VHEM-coarsened graphs have an effect of generating very efficient mappings, when they are compared to the HEM-coarsened graphs.

Original languageEnglish
Pages (from-to)530-548
Number of pages19
JournalJournal of Systems Architecture
Volume54
Issue number5
DOIs
Publication statusPublished - May 2008

Fingerprint

Coarsening
Tabu search
Distributed computer systems
Communication

Keywords

  • Heterogeneous computing systems
  • Mapping problem
  • Multilevel graph partitioning
  • Task allocation/assignment

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software

Cite this

@article{56a55efb928446839f20938c12db13d4,
title = "A multilevel partitioning approach for efficient tasks allocation in heterogeneous distributed systems",
abstract = "This work addresses the problem of allocating parallel application tasks to heterogeneous distributed computing resources, such as multiclusters or Grid environments. The proposed allocation scheme is based on a multilevel graph partitioning and mapping approach. The objective is to find an efficient allocation that minimizes the application completion time, subject to the specified constraints pertinent to the application and system environment. The allocation scheme consists of three phases; the clustering phase, the initial mapping phase and the refinement and remapping phase. The scheme introduces an efficient heuristic in the clustering phase for contracting (coarsening) large size application graphs to the number of processors, called the VHEM method. An initial mapping technique based on a tabu-search approach has been introduced as a basis for the process of refinement and remapping phase. The simulation study shows that the VHEM coarsening heuristic can achieve optimal or near-optimal communication, compared to the HEM method, when the ratio of the number of tasks to the number of processors exceeds a threshold value. The simulation study shows that those optimal or near-optimal VHEM-coarsened graphs have an effect of generating very efficient mappings, when they are compared to the HEM-coarsened graphs.",
keywords = "Heterogeneous computing systems, Mapping problem, Multilevel graph partitioning, Task allocation/assignment",
author = "Bassel Arafeh and Khalid Day and Abderezak Touzene",
year = "2008",
month = "5",
doi = "10.1016/j.sysarc.2007.10.001",
language = "English",
volume = "54",
pages = "530--548",
journal = "Journal of Systems Architecture",
issn = "1383-7621",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - A multilevel partitioning approach for efficient tasks allocation in heterogeneous distributed systems

AU - Arafeh, Bassel

AU - Day, Khalid

AU - Touzene, Abderezak

PY - 2008/5

Y1 - 2008/5

N2 - This work addresses the problem of allocating parallel application tasks to heterogeneous distributed computing resources, such as multiclusters or Grid environments. The proposed allocation scheme is based on a multilevel graph partitioning and mapping approach. The objective is to find an efficient allocation that minimizes the application completion time, subject to the specified constraints pertinent to the application and system environment. The allocation scheme consists of three phases; the clustering phase, the initial mapping phase and the refinement and remapping phase. The scheme introduces an efficient heuristic in the clustering phase for contracting (coarsening) large size application graphs to the number of processors, called the VHEM method. An initial mapping technique based on a tabu-search approach has been introduced as a basis for the process of refinement and remapping phase. The simulation study shows that the VHEM coarsening heuristic can achieve optimal or near-optimal communication, compared to the HEM method, when the ratio of the number of tasks to the number of processors exceeds a threshold value. The simulation study shows that those optimal or near-optimal VHEM-coarsened graphs have an effect of generating very efficient mappings, when they are compared to the HEM-coarsened graphs.

AB - This work addresses the problem of allocating parallel application tasks to heterogeneous distributed computing resources, such as multiclusters or Grid environments. The proposed allocation scheme is based on a multilevel graph partitioning and mapping approach. The objective is to find an efficient allocation that minimizes the application completion time, subject to the specified constraints pertinent to the application and system environment. The allocation scheme consists of three phases; the clustering phase, the initial mapping phase and the refinement and remapping phase. The scheme introduces an efficient heuristic in the clustering phase for contracting (coarsening) large size application graphs to the number of processors, called the VHEM method. An initial mapping technique based on a tabu-search approach has been introduced as a basis for the process of refinement and remapping phase. The simulation study shows that the VHEM coarsening heuristic can achieve optimal or near-optimal communication, compared to the HEM method, when the ratio of the number of tasks to the number of processors exceeds a threshold value. The simulation study shows that those optimal or near-optimal VHEM-coarsened graphs have an effect of generating very efficient mappings, when they are compared to the HEM-coarsened graphs.

KW - Heterogeneous computing systems

KW - Mapping problem

KW - Multilevel graph partitioning

KW - Task allocation/assignment

UR - http://www.scopus.com/inward/record.url?scp=43649091365&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=43649091365&partnerID=8YFLogxK

U2 - 10.1016/j.sysarc.2007.10.001

DO - 10.1016/j.sysarc.2007.10.001

M3 - Article

AN - SCOPUS:43649091365

VL - 54

SP - 530

EP - 548

JO - Journal of Systems Architecture

JF - Journal of Systems Architecture

SN - 1383-7621

IS - 5

ER -