Processor allocation and job scheduling on 3D mesh interconnection networks

S. B. Mohammad, M. O. Khaoua, L. M. Mackenzie, I. Ababneh

Research output: Contribution to journalArticle

Abstract

Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation, because it requires that the alio-cated processors be contiguous and have the same topology as the network topology connecting these processors. This paper suggests two non-contiguous processor allocation strategies, referred to as Paging and Greedy-Available respectively, for the 3D mesh network and compares their performance using software simulation against the well-known contiguous First Fit strategy. The comparative evaluation is conducted for two job scheduling strategies, notably first-come-first-served (FCFS) and out-of-order (OO). The results reveal that our proposed non-contiguous strategies exhibit superior performance properties despite the added contention that results from non-contiguity. The results also reveal that the scheduling and allocation strategies both have substantial effect on the performance of contiguous and non-contiguous allocation in that the OO scheduling strategy leads to much better performance than its FCFS counterpart in the 3D mesh.

Original languageEnglish
Pages (from-to)309-317
Number of pages9
JournalInternational Journal of Computers and Applications
Volume29
Issue number3
Publication statusPublished - 2007

Fingerprint

Scheduling
Topology

Keywords

  • Dispersal ratio
  • External message interference
  • Fragmentation
  • Multicomputer
  • Scheduling effectiveness
  • Turnaround time

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Processor allocation and job scheduling on 3D mesh interconnection networks. / Mohammad, S. B.; Khaoua, M. O.; Mackenzie, L. M.; Ababneh, I.

In: International Journal of Computers and Applications, Vol. 29, No. 3, 2007, p. 309-317.

Research output: Contribution to journalArticle

@article{7b1e29e7632847928f5e861e41d77bf2,
title = "Processor allocation and job scheduling on 3D mesh interconnection networks",
abstract = "Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation, because it requires that the alio-cated processors be contiguous and have the same topology as the network topology connecting these processors. This paper suggests two non-contiguous processor allocation strategies, referred to as Paging and Greedy-Available respectively, for the 3D mesh network and compares their performance using software simulation against the well-known contiguous First Fit strategy. The comparative evaluation is conducted for two job scheduling strategies, notably first-come-first-served (FCFS) and out-of-order (OO). The results reveal that our proposed non-contiguous strategies exhibit superior performance properties despite the added contention that results from non-contiguity. The results also reveal that the scheduling and allocation strategies both have substantial effect on the performance of contiguous and non-contiguous allocation in that the OO scheduling strategy leads to much better performance than its FCFS counterpart in the 3D mesh.",
keywords = "Dispersal ratio, External message interference, Fragmentation, Multicomputer, Scheduling effectiveness, Turnaround time",
author = "Mohammad, {S. B.} and Khaoua, {M. O.} and Mackenzie, {L. M.} and I. Ababneh",
year = "2007",
language = "English",
volume = "29",
pages = "309--317",
journal = "International Journal of Computers and Applications",
issn = "1206-212X",
publisher = "ACTA Press",
number = "3",

}

TY - JOUR

T1 - Processor allocation and job scheduling on 3D mesh interconnection networks

AU - Mohammad, S. B.

AU - Khaoua, M. O.

AU - Mackenzie, L. M.

AU - Ababneh, I.

PY - 2007

Y1 - 2007

N2 - Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation, because it requires that the alio-cated processors be contiguous and have the same topology as the network topology connecting these processors. This paper suggests two non-contiguous processor allocation strategies, referred to as Paging and Greedy-Available respectively, for the 3D mesh network and compares their performance using software simulation against the well-known contiguous First Fit strategy. The comparative evaluation is conducted for two job scheduling strategies, notably first-come-first-served (FCFS) and out-of-order (OO). The results reveal that our proposed non-contiguous strategies exhibit superior performance properties despite the added contention that results from non-contiguity. The results also reveal that the scheduling and allocation strategies both have substantial effect on the performance of contiguous and non-contiguous allocation in that the OO scheduling strategy leads to much better performance than its FCFS counterpart in the 3D mesh.

AB - Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation, because it requires that the alio-cated processors be contiguous and have the same topology as the network topology connecting these processors. This paper suggests two non-contiguous processor allocation strategies, referred to as Paging and Greedy-Available respectively, for the 3D mesh network and compares their performance using software simulation against the well-known contiguous First Fit strategy. The comparative evaluation is conducted for two job scheduling strategies, notably first-come-first-served (FCFS) and out-of-order (OO). The results reveal that our proposed non-contiguous strategies exhibit superior performance properties despite the added contention that results from non-contiguity. The results also reveal that the scheduling and allocation strategies both have substantial effect on the performance of contiguous and non-contiguous allocation in that the OO scheduling strategy leads to much better performance than its FCFS counterpart in the 3D mesh.

KW - Dispersal ratio

KW - External message interference

KW - Fragmentation

KW - Multicomputer

KW - Scheduling effectiveness

KW - Turnaround time

UR - http://www.scopus.com/inward/record.url?scp=35048894871&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048894871&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:35048894871

VL - 29

SP - 309

EP - 317

JO - International Journal of Computers and Applications

JF - International Journal of Computers and Applications

SN - 1206-212X

IS - 3

ER -