An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers

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

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

8 Citations (Scopus)

Abstract

Two strategies are Used for the allocation of jobs to processors connected by mesh topologies: contiguous allocation and non-contiguous allocation. In noncontiguous allocation, a job request can be split into smaller parts that are allocated to non-adjacent free submeshes rather than always waiting until a single submesh of the requested size and shape is available. Lifting the contiguity condition is expected to reduce processor fragmentation and increase system utilization. However, the distances traversed by messages can be long, and as a result the communication overhead, especially contention, is increased. The extra communication overhead depends on how the allocation request is partitioned and assigned to free sub-meshes. This paper presents a new Noncontiguous allocation algorithm, referred to as GreedyAvailable-Busy-List (GABL for short), which can decrease the communication overhead among processors allocated to a given job. The simulation results show that the new strategy can reduce the communication overhead and substantially improve performance in terms of parameters such as job turnaround time and system utilization. Moreover, the results reveal that the ShortestService-Demand-First (SSD) scheduling strategy is much better than the First-Come-First-Served (FCFS) scheduling strategy.

Original languageEnglish
Title of host publication2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007
Pages934-941
Number of pages8
DOIs
Publication statusPublished - 2007
Event2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007 - Amman, Jordan
Duration: May 13 2007May 16 2007

Other

Other2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007
CountryJordan
CityAmman
Period5/13/075/16/07

Fingerprint

Communication
Scheduling
Turnaround time
Topology

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Computer Science Applications
  • Hardware and Architecture
  • Signal Processing

Cite this

Bani-Mohammad, S., Ould-Khaoua, M., Ababneh, I., & Mackenzie, L. M. (2007). An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers. In 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007 (pp. 934-941). [4231071] https://doi.org/10.1109/AICCSA.2007.370743

An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers. / Bani-Mohammad, S.; Ould-Khaoua, M.; Ababneh, I.; Mackenzie, Lewis M.

2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007. 2007. p. 934-941 4231071.

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

Bani-Mohammad, S, Ould-Khaoua, M, Ababneh, I & Mackenzie, LM 2007, An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers. in 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007., 4231071, pp. 934-941, 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007, Amman, Jordan, 5/13/07. https://doi.org/10.1109/AICCSA.2007.370743
Bani-Mohammad S, Ould-Khaoua M, Ababneh I, Mackenzie LM. An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers. In 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007. 2007. p. 934-941. 4231071 https://doi.org/10.1109/AICCSA.2007.370743
Bani-Mohammad, S. ; Ould-Khaoua, M. ; Ababneh, I. ; Mackenzie, Lewis M. / An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers. 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007. 2007. pp. 934-941
@inproceedings{9048e9da701743cdb0e4f37acbeb52ef,
title = "An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers",
abstract = "Two strategies are Used for the allocation of jobs to processors connected by mesh topologies: contiguous allocation and non-contiguous allocation. In noncontiguous allocation, a job request can be split into smaller parts that are allocated to non-adjacent free submeshes rather than always waiting until a single submesh of the requested size and shape is available. Lifting the contiguity condition is expected to reduce processor fragmentation and increase system utilization. However, the distances traversed by messages can be long, and as a result the communication overhead, especially contention, is increased. The extra communication overhead depends on how the allocation request is partitioned and assigned to free sub-meshes. This paper presents a new Noncontiguous allocation algorithm, referred to as GreedyAvailable-Busy-List (GABL for short), which can decrease the communication overhead among processors allocated to a given job. The simulation results show that the new strategy can reduce the communication overhead and substantially improve performance in terms of parameters such as job turnaround time and system utilization. Moreover, the results reveal that the ShortestService-Demand-First (SSD) scheduling strategy is much better than the First-Come-First-Served (FCFS) scheduling strategy.",
author = "S. Bani-Mohammad and M. Ould-Khaoua and I. Ababneh and Mackenzie, {Lewis M.}",
year = "2007",
doi = "10.1109/AICCSA.2007.370743",
language = "English",
isbn = "1424410312",
pages = "934--941",
booktitle = "2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007",

}

TY - GEN

T1 - An efficient processor allocation strategy that maintains a high degree of contiguity among processors in 2D mesh connected multicomputers

AU - Bani-Mohammad, S.

AU - Ould-Khaoua, M.

AU - Ababneh, I.

AU - Mackenzie, Lewis M.

PY - 2007

Y1 - 2007

N2 - Two strategies are Used for the allocation of jobs to processors connected by mesh topologies: contiguous allocation and non-contiguous allocation. In noncontiguous allocation, a job request can be split into smaller parts that are allocated to non-adjacent free submeshes rather than always waiting until a single submesh of the requested size and shape is available. Lifting the contiguity condition is expected to reduce processor fragmentation and increase system utilization. However, the distances traversed by messages can be long, and as a result the communication overhead, especially contention, is increased. The extra communication overhead depends on how the allocation request is partitioned and assigned to free sub-meshes. This paper presents a new Noncontiguous allocation algorithm, referred to as GreedyAvailable-Busy-List (GABL for short), which can decrease the communication overhead among processors allocated to a given job. The simulation results show that the new strategy can reduce the communication overhead and substantially improve performance in terms of parameters such as job turnaround time and system utilization. Moreover, the results reveal that the ShortestService-Demand-First (SSD) scheduling strategy is much better than the First-Come-First-Served (FCFS) scheduling strategy.

AB - Two strategies are Used for the allocation of jobs to processors connected by mesh topologies: contiguous allocation and non-contiguous allocation. In noncontiguous allocation, a job request can be split into smaller parts that are allocated to non-adjacent free submeshes rather than always waiting until a single submesh of the requested size and shape is available. Lifting the contiguity condition is expected to reduce processor fragmentation and increase system utilization. However, the distances traversed by messages can be long, and as a result the communication overhead, especially contention, is increased. The extra communication overhead depends on how the allocation request is partitioned and assigned to free sub-meshes. This paper presents a new Noncontiguous allocation algorithm, referred to as GreedyAvailable-Busy-List (GABL for short), which can decrease the communication overhead among processors allocated to a given job. The simulation results show that the new strategy can reduce the communication overhead and substantially improve performance in terms of parameters such as job turnaround time and system utilization. Moreover, the results reveal that the ShortestService-Demand-First (SSD) scheduling strategy is much better than the First-Come-First-Served (FCFS) scheduling strategy.

UR - http://www.scopus.com/inward/record.url?scp=36249009021&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36249009021&partnerID=8YFLogxK

U2 - 10.1109/AICCSA.2007.370743

DO - 10.1109/AICCSA.2007.370743

M3 - Conference contribution

SN - 1424410312

SN - 9781424410316

SP - 934

EP - 941

BT - 2007 IEEE/ACS International Conference on Computer Systems and Applications, AICCSA 2007

ER -