Reliable workflow scheduling with less resource redundancy

Laiping Zhao, Yizhi Ren, Kouichi Sakurai

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We examine the problem of reliable workflow scheduling with less resource redundancy. As scheduling workflow applications in heterogeneous systems, either for optimizing the reliability or for minimizing the makespan, are NP-Complete problems, we alternatively find schedules for meeting specific reliability and deadline requirements. First, we analyze the reliability of a given schedule using two important definitions: Accumulated Processor Reliability (APR) and Accumulated Communication Reliability (ACR). Second, inspired by the reliability analysis, we present three scheduling algorithms: RR algorithm schedules least Resources to meet the Reliability requirement; DRR algorithm extends RR by further considering the Deadline requirement; and dynamic algorithm schedules tasks dynamically: It avoids the "Chain effect" caused by uncertainties on the task execution time estimates, and relieves the impact from the inaccuracy on failure estimation. Finally, the empirical evaluation shows that our algorithms can save a significant amount of computation and communication resources when performing a similar reliability compared to Fault-Tolerant-Scheduling-Algorithm (FTSA) algorithm.

Original languageEnglish
Pages (from-to)567-585
Number of pages19
JournalParallel Computing
Volume39
Issue number10
DOIs
Publication statusPublished - Jul 23 2013

Fingerprint

Work Flow
Redundancy
Scheduling
Resources
Schedule
Deadline
Scheduling algorithms
Scheduling Algorithm
Requirements
Dynamic Algorithms
Heterogeneous Systems
Communication
Reliability Analysis
Reliability analysis
Fault-tolerant
Execution Time
Computational complexity
NP-complete problem
Uncertainty
Evaluation

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computer Networks and Communications
  • Computer Graphics and Computer-Aided Design
  • Artificial Intelligence

Cite this

Reliable workflow scheduling with less resource redundancy. / Zhao, Laiping; Ren, Yizhi; Sakurai, Kouichi.

In: Parallel Computing, Vol. 39, No. 10, 23.07.2013, p. 567-585.

Research output: Contribution to journalArticle

Zhao, Laiping ; Ren, Yizhi ; Sakurai, Kouichi. / Reliable workflow scheduling with less resource redundancy. In: Parallel Computing. 2013 ; Vol. 39, No. 10. pp. 567-585.
@article{dc983525e8de4efb9e073e5197bd94aa,
title = "Reliable workflow scheduling with less resource redundancy",
abstract = "We examine the problem of reliable workflow scheduling with less resource redundancy. As scheduling workflow applications in heterogeneous systems, either for optimizing the reliability or for minimizing the makespan, are NP-Complete problems, we alternatively find schedules for meeting specific reliability and deadline requirements. First, we analyze the reliability of a given schedule using two important definitions: Accumulated Processor Reliability (APR) and Accumulated Communication Reliability (ACR). Second, inspired by the reliability analysis, we present three scheduling algorithms: RR algorithm schedules least Resources to meet the Reliability requirement; DRR algorithm extends RR by further considering the Deadline requirement; and dynamic algorithm schedules tasks dynamically: It avoids the {"}Chain effect{"} caused by uncertainties on the task execution time estimates, and relieves the impact from the inaccuracy on failure estimation. Finally, the empirical evaluation shows that our algorithms can save a significant amount of computation and communication resources when performing a similar reliability compared to Fault-Tolerant-Scheduling-Algorithm (FTSA) algorithm.",
author = "Laiping Zhao and Yizhi Ren and Kouichi Sakurai",
year = "2013",
month = "7",
day = "23",
doi = "10.1016/j.parco.2013.06.003",
language = "English",
volume = "39",
pages = "567--585",
journal = "Parallel Computing",
issn = "0167-8191",
publisher = "Elsevier",
number = "10",

}

TY - JOUR

T1 - Reliable workflow scheduling with less resource redundancy

AU - Zhao, Laiping

AU - Ren, Yizhi

AU - Sakurai, Kouichi

PY - 2013/7/23

Y1 - 2013/7/23

N2 - We examine the problem of reliable workflow scheduling with less resource redundancy. As scheduling workflow applications in heterogeneous systems, either for optimizing the reliability or for minimizing the makespan, are NP-Complete problems, we alternatively find schedules for meeting specific reliability and deadline requirements. First, we analyze the reliability of a given schedule using two important definitions: Accumulated Processor Reliability (APR) and Accumulated Communication Reliability (ACR). Second, inspired by the reliability analysis, we present three scheduling algorithms: RR algorithm schedules least Resources to meet the Reliability requirement; DRR algorithm extends RR by further considering the Deadline requirement; and dynamic algorithm schedules tasks dynamically: It avoids the "Chain effect" caused by uncertainties on the task execution time estimates, and relieves the impact from the inaccuracy on failure estimation. Finally, the empirical evaluation shows that our algorithms can save a significant amount of computation and communication resources when performing a similar reliability compared to Fault-Tolerant-Scheduling-Algorithm (FTSA) algorithm.

AB - We examine the problem of reliable workflow scheduling with less resource redundancy. As scheduling workflow applications in heterogeneous systems, either for optimizing the reliability or for minimizing the makespan, are NP-Complete problems, we alternatively find schedules for meeting specific reliability and deadline requirements. First, we analyze the reliability of a given schedule using two important definitions: Accumulated Processor Reliability (APR) and Accumulated Communication Reliability (ACR). Second, inspired by the reliability analysis, we present three scheduling algorithms: RR algorithm schedules least Resources to meet the Reliability requirement; DRR algorithm extends RR by further considering the Deadline requirement; and dynamic algorithm schedules tasks dynamically: It avoids the "Chain effect" caused by uncertainties on the task execution time estimates, and relieves the impact from the inaccuracy on failure estimation. Finally, the empirical evaluation shows that our algorithms can save a significant amount of computation and communication resources when performing a similar reliability compared to Fault-Tolerant-Scheduling-Algorithm (FTSA) algorithm.

UR - http://www.scopus.com/inward/record.url?scp=84884813091&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884813091&partnerID=8YFLogxK

U2 - 10.1016/j.parco.2013.06.003

DO - 10.1016/j.parco.2013.06.003

M3 - Article

AN - SCOPUS:84884813091

VL - 39

SP - 567

EP - 585

JO - Parallel Computing

JF - Parallel Computing

SN - 0167-8191

IS - 10

ER -