A task cluster size determination in heterogeneous distributed systems

Hidehiro Kanemitsu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

There are many task scheduling algorithms for minimizing the schedule length in distributed systems. Especially, task clustering heuristics are well known as an approach for deciding the number of required processors. However, the actual number may become very large depending on characteristics of the input job and each processor's capability. Thus, currently there is no theoretical criterion for achieving processor utilization. In this paper, we propose a method for deriving the lower bound of the cluster (assignment unit) size in heterogeneous distributed systems for limiting the number of processors. From results of experimental simulations, we discuss the applicability of the proposal to obtain better processor utilization.

Original languageEnglish
Title of host publicationInternational Conference on Advanced Communication Technology, ICACT
Pages1070-1075
Number of pages6
Publication statusPublished - 2012
Event14th International Conference on Advanced Communication Technology, ICACT 2012 - PyeongChang
Duration: 2012 Feb 192012 Feb 22

Other

Other14th International Conference on Advanced Communication Technology, ICACT 2012
CityPyeongChang
Period12/2/1912/2/22

Fingerprint

Size determination
Scheduling algorithms

Keywords

  • DAG
  • Heterogeneous Distributed Systems
  • Processor Utilization
  • Task Scheduling

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Kanemitsu, H. (2012). A task cluster size determination in heterogeneous distributed systems. In International Conference on Advanced Communication Technology, ICACT (pp. 1070-1075). [6174852]

A task cluster size determination in heterogeneous distributed systems. / Kanemitsu, Hidehiro.

International Conference on Advanced Communication Technology, ICACT. 2012. p. 1070-1075 6174852.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kanemitsu, H 2012, A task cluster size determination in heterogeneous distributed systems. in International Conference on Advanced Communication Technology, ICACT., 6174852, pp. 1070-1075, 14th International Conference on Advanced Communication Technology, ICACT 2012, PyeongChang, 12/2/19.
Kanemitsu H. A task cluster size determination in heterogeneous distributed systems. In International Conference on Advanced Communication Technology, ICACT. 2012. p. 1070-1075. 6174852
Kanemitsu, Hidehiro. / A task cluster size determination in heterogeneous distributed systems. International Conference on Advanced Communication Technology, ICACT. 2012. pp. 1070-1075
@inproceedings{77a0de49c0f44ff28a5f47259e878e11,
title = "A task cluster size determination in heterogeneous distributed systems",
abstract = "There are many task scheduling algorithms for minimizing the schedule length in distributed systems. Especially, task clustering heuristics are well known as an approach for deciding the number of required processors. However, the actual number may become very large depending on characteristics of the input job and each processor's capability. Thus, currently there is no theoretical criterion for achieving processor utilization. In this paper, we propose a method for deriving the lower bound of the cluster (assignment unit) size in heterogeneous distributed systems for limiting the number of processors. From results of experimental simulations, we discuss the applicability of the proposal to obtain better processor utilization.",
keywords = "DAG, Heterogeneous Distributed Systems, Processor Utilization, Task Scheduling",
author = "Hidehiro Kanemitsu",
year = "2012",
language = "English",
isbn = "9788955191639",
pages = "1070--1075",
booktitle = "International Conference on Advanced Communication Technology, ICACT",

}

TY - GEN

T1 - A task cluster size determination in heterogeneous distributed systems

AU - Kanemitsu, Hidehiro

PY - 2012

Y1 - 2012

N2 - There are many task scheduling algorithms for minimizing the schedule length in distributed systems. Especially, task clustering heuristics are well known as an approach for deciding the number of required processors. However, the actual number may become very large depending on characteristics of the input job and each processor's capability. Thus, currently there is no theoretical criterion for achieving processor utilization. In this paper, we propose a method for deriving the lower bound of the cluster (assignment unit) size in heterogeneous distributed systems for limiting the number of processors. From results of experimental simulations, we discuss the applicability of the proposal to obtain better processor utilization.

AB - There are many task scheduling algorithms for minimizing the schedule length in distributed systems. Especially, task clustering heuristics are well known as an approach for deciding the number of required processors. However, the actual number may become very large depending on characteristics of the input job and each processor's capability. Thus, currently there is no theoretical criterion for achieving processor utilization. In this paper, we propose a method for deriving the lower bound of the cluster (assignment unit) size in heterogeneous distributed systems for limiting the number of processors. From results of experimental simulations, we discuss the applicability of the proposal to obtain better processor utilization.

KW - DAG

KW - Heterogeneous Distributed Systems

KW - Processor Utilization

KW - Task Scheduling

UR - http://www.scopus.com/inward/record.url?scp=84860558263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860558263&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84860558263

SN - 9788955191639

SP - 1070

EP - 1075

BT - International Conference on Advanced Communication Technology, ICACT

ER -