Dispatching fixed-sized jobs with multiple deadlines to parallel heterogeneous servers

Esa Hyytiä; Rhonda Righter; Olivier Bilenne; Xiaohu Wu

doi:10.1016/j.peva.2017.04.003

Dispatching fixed-sized jobs with multiple deadlines to parallel heterogeneous servers

Esa Hyytiä^*, Rhonda Righter, Olivier Bilenne, Xiaohu Wu

^*Corresponding author for this work

Performance analysis

Research output: Contribution to journal › Article › Scientific › peer-review

6 Citations (Scopus)

188 Downloads (Pure)

Abstract

We study the M/D/1 queue when jobs have firm deadlines for waiting (or sojourn) time. If a deadline is not met, a job-specific deadline violation cost is incurred. We derive explicit value functions for this M/D/1 queue that enable the development of efficient cost-aware dispatching policies to parallel servers. The performance of the resulting dispatching policies is evaluated by means of simulations.

Original language	English
Pages (from-to)	32-44
Number of pages	13
Journal	Performance Evaluation
Volume	114
DOIs	https://doi.org/10.1016/j.peva.2017.04.003
Publication status	Published - 25 May 2017
MoE publication type	A1 Journal article-refereed

Keywords

Deadlines
Dispatching problem
M/D/1
MDP
Parallel computing

Access to Document

10.1016/j.peva.2017.04.003Licence: CC BY

1-s2.0-S0166531616302164-main
© 2017 The Authors. Published by Elsevier B.V. This is an open access article under the CC-BY Licence.
Final published version, 1.31 MBLicence: CC BY

Cite this

@article{61551c32c7c749a8b5338a552630b960,

title = "Dispatching fixed-sized jobs with multiple deadlines to parallel heterogeneous servers",

abstract = "We study the M/D/1 queue when jobs have firm deadlines for waiting (or sojourn) time. If a deadline is not met, a job-specific deadline violation cost is incurred. We derive explicit value functions for this M/D/1 queue that enable the development of efficient cost-aware dispatching policies to parallel servers. The performance of the resulting dispatching policies is evaluated by means of simulations.",

keywords = "Deadlines, Dispatching problem, M/D/1, MDP, Parallel computing",

author = "Esa Hyyti{\"a} and Rhonda Righter and Olivier Bilenne and Xiaohu Wu",

year = "2017",

month = may,

day = "25",

doi = "10.1016/j.peva.2017.04.003",

language = "English",

volume = "114",

pages = "32--44",

journal = "Performance Evaluation",

issn = "0166-5316",

publisher = "Elsevier",

}

TY - JOUR

T1 - Dispatching fixed-sized jobs with multiple deadlines to parallel heterogeneous servers

AU - Hyytiä, Esa

AU - Righter, Rhonda

AU - Bilenne, Olivier

AU - Wu, Xiaohu

PY - 2017/5/25

Y1 - 2017/5/25

N2 - We study the M/D/1 queue when jobs have firm deadlines for waiting (or sojourn) time. If a deadline is not met, a job-specific deadline violation cost is incurred. We derive explicit value functions for this M/D/1 queue that enable the development of efficient cost-aware dispatching policies to parallel servers. The performance of the resulting dispatching policies is evaluated by means of simulations.

AB - We study the M/D/1 queue when jobs have firm deadlines for waiting (or sojourn) time. If a deadline is not met, a job-specific deadline violation cost is incurred. We derive explicit value functions for this M/D/1 queue that enable the development of efficient cost-aware dispatching policies to parallel servers. The performance of the resulting dispatching policies is evaluated by means of simulations.

KW - Deadlines

KW - Dispatching problem

KW - M/D/1

KW - MDP

KW - Parallel computing

UR - http://www.scopus.com/inward/record.url?scp=85020903652&partnerID=8YFLogxK

U2 - 10.1016/j.peva.2017.04.003

DO - 10.1016/j.peva.2017.04.003

M3 - Article

AN - SCOPUS:85020903652

SN - 0166-5316

VL - 114

SP - 32

EP - 44

JO - Performance Evaluation

JF - Performance Evaluation

ER -

Dispatching fixed-sized jobs with multiple deadlines to parallel heterogeneous servers

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this