Approximate performance analysis of generalized join the shortest queue routing

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

1 Citaat (Scopus)

Uittreksel

In this paper we propose a highly accurate approximate performance analysis of a heterogeneous server system with a processor sharing service discipline and a general job-size distribution under a generalized join the shortest queue (GJSQ) routing protocol. The GJSQ routing protocol is a natural extension of the well-known join the shortest queue routing policy that takes into account the non-identical service rates in addition to the number of jobs at each server. The performance metrics that are of interest here are the equilibrium distribution and the mean and standard deviation of the number of jobs at each server. We show that the latter metrics are near-insensitive to the job-size distribution using simulation experiments. By applying a single queue approximation we model each server as a single server queue with a state-dependent arrival process, independent of other servers in the system, and derive the distribution of the number of jobs at the server. These state-dependent arrival rates are intended to capture the inherent correlation between servers in the original system and behave in a rather atypical way
TaalEngels
TitelVALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015
RedacteurenW. Knottenbelt, K. Wolter, A. Busic, M. Gribaudo, P. Reinecke
Plaats van productieBrussels, Belgium
UitgeverijInstitute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST)
Pagina's103-110
Aantal pagina's8
ISBN van elektronische versie978-1-63190-096-9
DOI's
StatusGepubliceerd - 2016
Evenement9th EAI International Conference on Performance Evaluation Methodologies and Tools (Valuetools 2015), December 14–16, 2015, Berlin, Germany - Berlin, Duitsland
Duur: 14 dec 201516 dec 2015
http://valuetools.org/2015/show/home

Congres

Congres9th EAI International Conference on Performance Evaluation Methodologies and Tools (Valuetools 2015), December 14–16, 2015, Berlin, Germany
Verkorte titelValuetools 2015
LandDuitsland
StadBerlin
Periode14/12/1516/12/15
Internet adres

Vingerafdruk

Queue
Join
Routing
Performance analysis
Simulation experiment
Approximation
Equilibrium distribution
Standard deviation
Performance metrics

Citeer dit

Selen, J., Adan, I. J. B. F., & Kapodistria, S. (2016). Approximate performance analysis of generalized join the shortest queue routing. In W. Knottenbelt, K. Wolter, A. Busic, M. Gribaudo, & P. Reinecke (editors), VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015 (blz. 103-110). Brussels, Belgium: Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST). DOI: 10.4108/eai.14-12-2015.2262695
Selen, J. ; Adan, I.J.B.F. ; Kapodistria, S./ Approximate performance analysis of generalized join the shortest queue routing. VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015. redacteur / W. Knottenbelt ; K. Wolter ; A. Busic ; M. Gribaudo ; P. Reinecke. Brussels, Belgium : Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST), 2016. blz. 103-110
@inproceedings{53b235d29b5748deab352fb1ab730c46,
title = "Approximate performance analysis of generalized join the shortest queue routing",
abstract = "In this paper we propose a highly accurate approximate performance analysis of a heterogeneous server system with a processor sharing service discipline and a general job-size distribution under a generalized join the shortest queue (GJSQ) routing protocol. The GJSQ routing protocol is a natural extension of the well-known join the shortest queue routing policy that takes into account the non-identical service rates in addition to the number of jobs at each server. The performance metrics that are of interest here are the equilibrium distribution and the mean and standard deviation of the number of jobs at each server. We show that the latter metrics are near-insensitive to the job-size distribution using simulation experiments. By applying a single queue approximation we model each server as a single server queue with a state-dependent arrival process, independent of other servers in the system, and derive the distribution of the number of jobs at the server. These state-dependent arrival rates are intended to capture the inherent correlation between servers in the original system and behave in a rather atypical way",
author = "J. Selen and I.J.B.F. Adan and S. Kapodistria",
year = "2016",
doi = "10.4108/eai.14-12-2015.2262695",
language = "English",
pages = "103--110",
editor = "W. Knottenbelt and K. Wolter and A. Busic and M. Gribaudo and P. Reinecke",
booktitle = "VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015",
publisher = "Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST)",

}

Selen, J, Adan, IJBF & Kapodistria, S 2016, Approximate performance analysis of generalized join the shortest queue routing. in W Knottenbelt, K Wolter, A Busic, M Gribaudo & P Reinecke (redactie), VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015. Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST), Brussels, Belgium, blz. 103-110, Berlin, Duitsland, 14/12/15. DOI: 10.4108/eai.14-12-2015.2262695

Approximate performance analysis of generalized join the shortest queue routing. / Selen, J.; Adan, I.J.B.F.; Kapodistria, S.

VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015. redactie / W. Knottenbelt; K. Wolter; A. Busic; M. Gribaudo; P. Reinecke. Brussels, Belgium : Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST), 2016. blz. 103-110.

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

TY - GEN

T1 - Approximate performance analysis of generalized join the shortest queue routing

AU - Selen,J.

AU - Adan,I.J.B.F.

AU - Kapodistria,S.

PY - 2016

Y1 - 2016

N2 - In this paper we propose a highly accurate approximate performance analysis of a heterogeneous server system with a processor sharing service discipline and a general job-size distribution under a generalized join the shortest queue (GJSQ) routing protocol. The GJSQ routing protocol is a natural extension of the well-known join the shortest queue routing policy that takes into account the non-identical service rates in addition to the number of jobs at each server. The performance metrics that are of interest here are the equilibrium distribution and the mean and standard deviation of the number of jobs at each server. We show that the latter metrics are near-insensitive to the job-size distribution using simulation experiments. By applying a single queue approximation we model each server as a single server queue with a state-dependent arrival process, independent of other servers in the system, and derive the distribution of the number of jobs at the server. These state-dependent arrival rates are intended to capture the inherent correlation between servers in the original system and behave in a rather atypical way

AB - In this paper we propose a highly accurate approximate performance analysis of a heterogeneous server system with a processor sharing service discipline and a general job-size distribution under a generalized join the shortest queue (GJSQ) routing protocol. The GJSQ routing protocol is a natural extension of the well-known join the shortest queue routing policy that takes into account the non-identical service rates in addition to the number of jobs at each server. The performance metrics that are of interest here are the equilibrium distribution and the mean and standard deviation of the number of jobs at each server. We show that the latter metrics are near-insensitive to the job-size distribution using simulation experiments. By applying a single queue approximation we model each server as a single server queue with a state-dependent arrival process, independent of other servers in the system, and derive the distribution of the number of jobs at the server. These state-dependent arrival rates are intended to capture the inherent correlation between servers in the original system and behave in a rather atypical way

U2 - 10.4108/eai.14-12-2015.2262695

DO - 10.4108/eai.14-12-2015.2262695

M3 - Conference contribution

SP - 103

EP - 110

BT - VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015

PB - Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST)

CY - Brussels, Belgium

ER -

Selen J, Adan IJBF, Kapodistria S. Approximate performance analysis of generalized join the shortest queue routing. In Knottenbelt W, Wolter K, Busic A, Gribaudo M, Reinecke P, redacteurs, VALUETOOLS'15 Proceedings of the 9th EAI International Conference on Performance Evaluation Methodologies and Tools, Berlin, 14-16 December 2015. Brussels, Belgium: Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (ICST). 2016. blz. 103-110. Beschikbaar vanaf, DOI: 10.4108/eai.14-12-2015.2262695