Feedback-based admission control for task allocation

Piotr Dziurzanski; Hashem Ali Ghazzawi; Leandro Soares Indrusiak

doi:10.1109/ReCoSoC.2014.6861364

Feedback-based admission control for task allocation

Piotr Dziurzanski, Hashem Ali Ghazzawi, Leandro Soares Indrusiak

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

This paper presents early exploration of the feedback admission control for high-performance computing clusters executing real-time tasks using controlled values to dynamic task allocation. A number of controller variants with different architectures and relying on various metrics have been proposed. Simulation models of both open- and closed-loop systems have been prepared and compared. The obtained experimental results show that the proposed approach leads to executing almost five times more tasks before their deadlines in case of periodic uniform workload, and about 16% more for bursty workloads with large computation time variance.

Original language	English
Title of host publication	Proceedings of the 9th International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC)
Publisher	IEEE Computer Society
ISBN (Print)	9781479958108
DOIs	https://doi.org/10.1109/ReCoSoC.2014.6861364
Publication status	Published - 2014
Event	9th International Symposium on Reconfigurable Communication-centric System-on-Chip (ReCoSoC) - Montpellier, France Duration: 26 May 2014 → 28 May 2014

Conference

Conference	9th International Symposium on Reconfigurable Communication-centric System-on-Chip (ReCoSoC)
Country/Territory	France
City	Montpellier
Period	26/05/14 → 28/05/14

Access to Document

10.1109/ReCoSoC.2014.6861364

Cite this

@inproceedings{1bd5d54c869e4d7bafa9001d1d33611f,

title = "Feedback-based admission control for task allocation",

abstract = "This paper presents early exploration of the feedback admission control for high-performance computing clusters executing real-time tasks using controlled values to dynamic task allocation. A number of controller variants with different architectures and relying on various metrics have been proposed. Simulation models of both open- and closed-loop systems have been prepared and compared. The obtained experimental results show that the proposed approach leads to executing almost five times more tasks before their deadlines in case of periodic uniform workload, and about 16% more for bursty workloads with large computation time variance.",

author = "Piotr Dziurzanski and Ghazzawi, {Hashem Ali} and Indrusiak, {Leandro Soares}",

year = "2014",

doi = "10.1109/ReCoSoC.2014.6861364",

language = "English",

isbn = "9781479958108",

booktitle = "Proceedings of the 9th International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC)",

publisher = "IEEE Computer Society",

address = "United States",

note = "9th International Symposium on Reconfigurable Communication-centric System-on-Chip (ReCoSoC) ; Conference date: 26-05-2014 Through 28-05-2014",

}

Dziurzanski, P, Ghazzawi, HA & Indrusiak, LS 2014, Feedback-based admission control for task allocation. in Proceedings of the 9th International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC). IEEE Computer Society, 9th International Symposium on Reconfigurable Communication-centric System-on-Chip (ReCoSoC), Montpellier, France, 26/05/14. https://doi.org/10.1109/ReCoSoC.2014.6861364

TY - GEN

T1 - Feedback-based admission control for task allocation

AU - Dziurzanski, Piotr

AU - Ghazzawi, Hashem Ali

AU - Indrusiak, Leandro Soares

PY - 2014

Y1 - 2014

N2 - This paper presents early exploration of the feedback admission control for high-performance computing clusters executing real-time tasks using controlled values to dynamic task allocation. A number of controller variants with different architectures and relying on various metrics have been proposed. Simulation models of both open- and closed-loop systems have been prepared and compared. The obtained experimental results show that the proposed approach leads to executing almost five times more tasks before their deadlines in case of periodic uniform workload, and about 16% more for bursty workloads with large computation time variance.

AB - This paper presents early exploration of the feedback admission control for high-performance computing clusters executing real-time tasks using controlled values to dynamic task allocation. A number of controller variants with different architectures and relying on various metrics have been proposed. Simulation models of both open- and closed-loop systems have been prepared and compared. The obtained experimental results show that the proposed approach leads to executing almost five times more tasks before their deadlines in case of periodic uniform workload, and about 16% more for bursty workloads with large computation time variance.

UR - http://www.scopus.com/inward/record.url?scp=84905638758&partnerID=8YFLogxK

U2 - 10.1109/ReCoSoC.2014.6861364

DO - 10.1109/ReCoSoC.2014.6861364

M3 - Conference contribution

AN - SCOPUS:84905638758

SN - 9781479958108

BT - Proceedings of the 9th International Symposium on Reconfigurable and Communication-Centric Systems-on-Chip (ReCoSoC)

PB - IEEE Computer Society

T2 - 9th International Symposium on Reconfigurable Communication-centric System-on-Chip (ReCoSoC)

Y2 - 26 May 2014 through 28 May 2014

ER -

Feedback-based admission control for task allocation

Abstract

Conference

Access to Document

Other files and links

DreamCloud

Cite this

Feedback-based admission control for task allocation

Abstract

Conference

Access to Document

Other files and links

Projects

DreamCloud

Cite this