Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation

S. Bani-Mohammad, M. Ould-Khaoua, I. Ababneh, Lewis M. Mackenzie

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

21 Citations (Scopus)

Abstract

Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation as it requires that the allocated processors be contiguous and has the same topology as the network topology connecting these processors. In non-contiguous allocation, a job can execute on multiple disjoint smaller sub-meshes rather than always waiting until a single sub-mesh of the requested size is available. Lifting the contiguity condition in non-contiguous allocation is expected to reduce processor fragmentation and increase processor utilization. However, the communication overhead is increased because the distances traversed by messages can be longer. The extra communication overhead depends on how the allocation request is partitioned and allocated to free sub-meshes. In this paper, a new non-contiguous processor allocation strategy, referred to as Greedy-Available-Busy-List, is suggested for the 2D mesh network, and is compared using simulation against the well-known non-contiguous and contiguous allocation strategies. To show the performance improved by proposed strategy, we conducted simulation runs under the assumption of wormhole routing and all-to-all communication pattern. The results show that the proposed strategy can reduce the communication overhead and improve performance substantially in terms of turnaround times of jobs and finish times.

Original languageEnglish
Title of host publicationProceedings of the International Conference on Parallel and Distributed Systems - ICPADS
Pages41-48
Number of pages8
Volume2
DOIs
Publication statusPublished - 2006
Event12th International Conference on Parallel and Distributed Systems, ICPADS 2006 - Minneapolis, MN, United States
Duration: Jul 12 2006Jul 15 2006

Other

Other12th International Conference on Parallel and Distributed Systems, ICPADS 2006
CountryUnited States
CityMinneapolis, MN
Period7/12/067/15/06

Fingerprint

Communication
Topology
Turnaround time

Keywords

  • Contiguous allocation
  • Finish time
  • Fragmentation
  • Multicomputers
  • Non-contigous allocation
  • Performance Comparison
  • Simulation
  • System utilization
  • Turnaround time

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Bani-Mohammad, S., Ould-Khaoua, M., Ababneh, I., & Mackenzie, L. M. (2006). Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation. In Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS (Vol. 2, pp. 41-48). [1654600] https://doi.org/10.1109/ICPADS.2006.68

Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation. / Bani-Mohammad, S.; Ould-Khaoua, M.; Ababneh, I.; Mackenzie, Lewis M.

Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS. Vol. 2 2006. p. 41-48 1654600.

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

Bani-Mohammad, S, Ould-Khaoua, M, Ababneh, I & Mackenzie, LM 2006, Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation. in Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS. vol. 2, 1654600, pp. 41-48, 12th International Conference on Parallel and Distributed Systems, ICPADS 2006, Minneapolis, MN, United States, 7/12/06. https://doi.org/10.1109/ICPADS.2006.68
Bani-Mohammad S, Ould-Khaoua M, Ababneh I, Mackenzie LM. Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation. In Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS. Vol. 2. 2006. p. 41-48. 1654600 https://doi.org/10.1109/ICPADS.2006.68
Bani-Mohammad, S. ; Ould-Khaoua, M. ; Ababneh, I. ; Mackenzie, Lewis M. / Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation. Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS. Vol. 2 2006. pp. 41-48
@inproceedings{6a6421cc3e7f497e9031b352b636cb71,
title = "Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation",
abstract = "Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation as it requires that the allocated processors be contiguous and has the same topology as the network topology connecting these processors. In non-contiguous allocation, a job can execute on multiple disjoint smaller sub-meshes rather than always waiting until a single sub-mesh of the requested size is available. Lifting the contiguity condition in non-contiguous allocation is expected to reduce processor fragmentation and increase processor utilization. However, the communication overhead is increased because the distances traversed by messages can be longer. The extra communication overhead depends on how the allocation request is partitioned and allocated to free sub-meshes. In this paper, a new non-contiguous processor allocation strategy, referred to as Greedy-Available-Busy-List, is suggested for the 2D mesh network, and is compared using simulation against the well-known non-contiguous and contiguous allocation strategies. To show the performance improved by proposed strategy, we conducted simulation runs under the assumption of wormhole routing and all-to-all communication pattern. The results show that the proposed strategy can reduce the communication overhead and improve performance substantially in terms of turnaround times of jobs and finish times.",
keywords = "Contiguous allocation, Finish time, Fragmentation, Multicomputers, Non-contigous allocation, Performance Comparison, Simulation, System utilization, Turnaround time",
author = "S. Bani-Mohammad and M. Ould-Khaoua and I. Ababneh and Mackenzie, {Lewis M.}",
year = "2006",
doi = "10.1109/ICPADS.2006.68",
language = "English",
isbn = "0769526128",
volume = "2",
pages = "41--48",
booktitle = "Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS",

}

TY - GEN

T1 - Non-contiguous processor allocation strategy for 2D mesh connected multicomputers based on sub-meshes available for allocation

AU - Bani-Mohammad, S.

AU - Ould-Khaoua, M.

AU - Ababneh, I.

AU - Mackenzie, Lewis M.

PY - 2006

Y1 - 2006

N2 - Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation as it requires that the allocated processors be contiguous and has the same topology as the network topology connecting these processors. In non-contiguous allocation, a job can execute on multiple disjoint smaller sub-meshes rather than always waiting until a single sub-mesh of the requested size is available. Lifting the contiguity condition in non-contiguous allocation is expected to reduce processor fragmentation and increase processor utilization. However, the communication overhead is increased because the distances traversed by messages can be longer. The extra communication overhead depends on how the allocation request is partitioned and allocated to free sub-meshes. In this paper, a new non-contiguous processor allocation strategy, referred to as Greedy-Available-Busy-List, is suggested for the 2D mesh network, and is compared using simulation against the well-known non-contiguous and contiguous allocation strategies. To show the performance improved by proposed strategy, we conducted simulation runs under the assumption of wormhole routing and all-to-all communication pattern. The results show that the proposed strategy can reduce the communication overhead and improve performance substantially in terms of turnaround times of jobs and finish times.

AB - Contiguous allocation of parallel jobs usually suffers from the degrading effects of fragmentation as it requires that the allocated processors be contiguous and has the same topology as the network topology connecting these processors. In non-contiguous allocation, a job can execute on multiple disjoint smaller sub-meshes rather than always waiting until a single sub-mesh of the requested size is available. Lifting the contiguity condition in non-contiguous allocation is expected to reduce processor fragmentation and increase processor utilization. However, the communication overhead is increased because the distances traversed by messages can be longer. The extra communication overhead depends on how the allocation request is partitioned and allocated to free sub-meshes. In this paper, a new non-contiguous processor allocation strategy, referred to as Greedy-Available-Busy-List, is suggested for the 2D mesh network, and is compared using simulation against the well-known non-contiguous and contiguous allocation strategies. To show the performance improved by proposed strategy, we conducted simulation runs under the assumption of wormhole routing and all-to-all communication pattern. The results show that the proposed strategy can reduce the communication overhead and improve performance substantially in terms of turnaround times of jobs and finish times.

KW - Contiguous allocation

KW - Finish time

KW - Fragmentation

KW - Multicomputers

KW - Non-contigous allocation

KW - Performance Comparison

KW - Simulation

KW - System utilization

KW - Turnaround time

UR - http://www.scopus.com/inward/record.url?scp=33947407404&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947407404&partnerID=8YFLogxK

U2 - 10.1109/ICPADS.2006.68

DO - 10.1109/ICPADS.2006.68

M3 - Conference contribution

SN - 0769526128

SN - 9780769526126

VL - 2

SP - 41

EP - 48

BT - Proceedings of the International Conference on Parallel and Distributed Systems - ICPADS

ER -