A simulated evolution approach to task matching and scheduling in heterogeneous computing environments

Hassan Barada; Sadiq M. Sait; Naved Baig

doi:10.1016/S0952-1976(02)00088-X

A simulated evolution approach to task matching and scheduling in heterogeneous computing environments

Hassan Barada
, Sadiq M. Sait
, Naved Baig

Electrical Engineering

King Fahd University for Petroleum and Minerals (KFUPM)

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

This paper applies a simulated evolution (SE) approach to the problem of matching and scheduling dependent tasks in a heterogeneous suite of computers interconnected via a high-speed network. The various steps of the SE approach are discussed in details. Goodness functions required by SE are designed and explained. Experimental results applied on various types of workloads are analyzed. Workloads are characterized according to the connectivity, heterogeneity, and communication-to-cost ratio of the task graphs representing the application tasks. The performance of SE is compared with a genetic algorithm approach for the same problem with respect to the quality of solutions generated, and timing requirements of the algorithms.

Original language	British English
Pages (from-to)	491-500
Number of pages	10
Journal	Engineering Applications of Artificial Intelligence
Volume	15
Issue number	5
DOIs	https://doi.org/10.1016/S0952-1976(02)00088-X
State	Published - Sep 2002

Keywords

Heterogeneous systems
Heuristics
Optimization
Scheduling

Access to Document

10.1016/S0952-1976(02)00088-X

Cite this

@article{2c7a5291eea1476caf87f289e65a5454,

title = "A simulated evolution approach to task matching and scheduling in heterogeneous computing environments",

abstract = "This paper applies a simulated evolution (SE) approach to the problem of matching and scheduling dependent tasks in a heterogeneous suite of computers interconnected via a high-speed network. The various steps of the SE approach are discussed in details. Goodness functions required by SE are designed and explained. Experimental results applied on various types of workloads are analyzed. Workloads are characterized according to the connectivity, heterogeneity, and communication-to-cost ratio of the task graphs representing the application tasks. The performance of SE is compared with a genetic algorithm approach for the same problem with respect to the quality of solutions generated, and timing requirements of the algorithms.",

keywords = "Heterogeneous systems, Heuristics, Optimization, Scheduling",

author = "Hassan Barada and Sait, \{Sadiq M.\} and Naved Baig",

year = "2002",

month = sep,

doi = "10.1016/S0952-1976(02)00088-X",

language = "British English",

volume = "15",

pages = "491--500",

journal = "Engineering Applications of Artificial Intelligence",

issn = "0952-1976",

publisher = "Elsevier Ltd",

number = "5",

}

TY - JOUR

T1 - A simulated evolution approach to task matching and scheduling in heterogeneous computing environments

AU - Barada, Hassan

AU - Sait, Sadiq M.

AU - Baig, Naved

PY - 2002/9

Y1 - 2002/9

N2 - This paper applies a simulated evolution (SE) approach to the problem of matching and scheduling dependent tasks in a heterogeneous suite of computers interconnected via a high-speed network. The various steps of the SE approach are discussed in details. Goodness functions required by SE are designed and explained. Experimental results applied on various types of workloads are analyzed. Workloads are characterized according to the connectivity, heterogeneity, and communication-to-cost ratio of the task graphs representing the application tasks. The performance of SE is compared with a genetic algorithm approach for the same problem with respect to the quality of solutions generated, and timing requirements of the algorithms.

AB - This paper applies a simulated evolution (SE) approach to the problem of matching and scheduling dependent tasks in a heterogeneous suite of computers interconnected via a high-speed network. The various steps of the SE approach are discussed in details. Goodness functions required by SE are designed and explained. Experimental results applied on various types of workloads are analyzed. Workloads are characterized according to the connectivity, heterogeneity, and communication-to-cost ratio of the task graphs representing the application tasks. The performance of SE is compared with a genetic algorithm approach for the same problem with respect to the quality of solutions generated, and timing requirements of the algorithms.

KW - Heterogeneous systems

KW - Heuristics

KW - Optimization

KW - Scheduling

UR - https://www.scopus.com/pages/publications/0036771094

U2 - 10.1016/S0952-1976(02)00088-X

DO - 10.1016/S0952-1976(02)00088-X

M3 - Article

AN - SCOPUS:0036771094

SN - 0952-1976

VL - 15

SP - 491

EP - 500

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

IS - 5

ER -

A simulated evolution approach to task matching and scheduling in heterogeneous computing environments

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this