High-throughput scientific workflow scheduling under deadline constraint in clouds

Michelle Zhu, Fei Cao, Chase Q. Wu

Research output: Contribution to journalArticleResearchpeer-review

5 Citations (Scopus)

Abstract

Cloud computing is a paradigm shift in service delivery that promises a leap in efficiency and flexibility in using computing resources. As cloud infrastructures are widely deployed around the globe, many data- and computeintensive scientific workflows have been moved from traditional high-performance computing platforms and grids to clouds. With the rapidly increasing number of cloud users in various science domains, it has become a critical task for the cloud service provider to perform efficient job scheduling while still guaranteeing the workflow completion time as specified in the Service Level Agreement (SLA). Based on practical models for cloud utilization, we formulate a delay-constrained workflow optimization problem to maximize resource utilization for high system throughput and propose a two-step scheduling algorithm to minimize the cloud overhead under a user-specified execution time bound. Extensive simulation results illustrate that the proposed algorithm achieves lower computing overhead or higher resource utilization than existing methods under the execution time bound, and also significantly reduces the total workflow execution time by strategically selecting appropriate mapping nodes for prioritized modules.

Original languageEnglish
Pages (from-to)312-321
Number of pages10
JournalJournal of Communications
Volume9
Issue number4
DOIs
StatePublished - 1 Jan 2014

Fingerprint

Constrained optimization
Cloud computing
Scheduling algorithms
Scheduling
Throughput

Keywords

  • Cloudcomputing
  • Scientific workflow
  • Workflow scheduling

Cite this

@article{49ba6176fc594826ae7d62bdd83dc177,
title = "High-throughput scientific workflow scheduling under deadline constraint in clouds",
abstract = "Cloud computing is a paradigm shift in service delivery that promises a leap in efficiency and flexibility in using computing resources. As cloud infrastructures are widely deployed around the globe, many data- and computeintensive scientific workflows have been moved from traditional high-performance computing platforms and grids to clouds. With the rapidly increasing number of cloud users in various science domains, it has become a critical task for the cloud service provider to perform efficient job scheduling while still guaranteeing the workflow completion time as specified in the Service Level Agreement (SLA). Based on practical models for cloud utilization, we formulate a delay-constrained workflow optimization problem to maximize resource utilization for high system throughput and propose a two-step scheduling algorithm to minimize the cloud overhead under a user-specified execution time bound. Extensive simulation results illustrate that the proposed algorithm achieves lower computing overhead or higher resource utilization than existing methods under the execution time bound, and also significantly reduces the total workflow execution time by strategically selecting appropriate mapping nodes for prioritized modules.",
keywords = "Cloudcomputing, Scientific workflow, Workflow scheduling",
author = "Michelle Zhu and Fei Cao and Wu, {Chase Q.}",
year = "2014",
month = "1",
day = "1",
doi = "10.12720/jcm.9.4.312-321",
language = "English",
volume = "9",
pages = "312--321",
journal = "Journal of Communications",
issn = "1796-2021",
publisher = "Engineering and Technology Publishing",
number = "4",

}

High-throughput scientific workflow scheduling under deadline constraint in clouds. / Zhu, Michelle; Cao, Fei; Wu, Chase Q.

In: Journal of Communications, Vol. 9, No. 4, 01.01.2014, p. 312-321.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - High-throughput scientific workflow scheduling under deadline constraint in clouds

AU - Zhu, Michelle

AU - Cao, Fei

AU - Wu, Chase Q.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Cloud computing is a paradigm shift in service delivery that promises a leap in efficiency and flexibility in using computing resources. As cloud infrastructures are widely deployed around the globe, many data- and computeintensive scientific workflows have been moved from traditional high-performance computing platforms and grids to clouds. With the rapidly increasing number of cloud users in various science domains, it has become a critical task for the cloud service provider to perform efficient job scheduling while still guaranteeing the workflow completion time as specified in the Service Level Agreement (SLA). Based on practical models for cloud utilization, we formulate a delay-constrained workflow optimization problem to maximize resource utilization for high system throughput and propose a two-step scheduling algorithm to minimize the cloud overhead under a user-specified execution time bound. Extensive simulation results illustrate that the proposed algorithm achieves lower computing overhead or higher resource utilization than existing methods under the execution time bound, and also significantly reduces the total workflow execution time by strategically selecting appropriate mapping nodes for prioritized modules.

AB - Cloud computing is a paradigm shift in service delivery that promises a leap in efficiency and flexibility in using computing resources. As cloud infrastructures are widely deployed around the globe, many data- and computeintensive scientific workflows have been moved from traditional high-performance computing platforms and grids to clouds. With the rapidly increasing number of cloud users in various science domains, it has become a critical task for the cloud service provider to perform efficient job scheduling while still guaranteeing the workflow completion time as specified in the Service Level Agreement (SLA). Based on practical models for cloud utilization, we formulate a delay-constrained workflow optimization problem to maximize resource utilization for high system throughput and propose a two-step scheduling algorithm to minimize the cloud overhead under a user-specified execution time bound. Extensive simulation results illustrate that the proposed algorithm achieves lower computing overhead or higher resource utilization than existing methods under the execution time bound, and also significantly reduces the total workflow execution time by strategically selecting appropriate mapping nodes for prioritized modules.

KW - Cloudcomputing

KW - Scientific workflow

KW - Workflow scheduling

UR - http://www.scopus.com/inward/record.url?scp=84899696483&partnerID=8YFLogxK

U2 - 10.12720/jcm.9.4.312-321

DO - 10.12720/jcm.9.4.312-321

M3 - Article

VL - 9

SP - 312

EP - 321

JO - Journal of Communications

JF - Journal of Communications

SN - 1796-2021

IS - 4

ER -