Cloudde: a heterogeneous differential evolution algorithm and its distributed cloud version

Zhi-Hui Zhan, Xiao-Fang Liu, Huaxiang Zhang, Zhengtao Yu, Jian Weng, Yun Li, Tianlong Gu, Jun Zhang

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

Existing differential evolution (DE) algorithms often face two challenges. The first is that the optimization performance is significantly affected by the ad hoc configurations of operators and parameters for different problems. The second is the long runtime for real-world problems whose fitness evaluations are often expensive. Aiming at solving these two problems, this paper develops a novel double-layered heterogeneous DE algorithm and realizes it in cloud computing distributed environment. In the first layer, different populations with various parameters and/or operators run concurrently and adaptively migrate to deliver robust solutions by making the best use of performance differences among multiple populations. In the second layer, a set of cloud virtual machines run in parallel to evaluate fitness of corresponding populations, reducing computational costs as offered by cloud. Experimental results on a set of benchmark problems with different search requirements and a case study with expensive design evaluations have shown that the proposed algorithm offers generally improved performance and reduced computational time, compared with not only conventional and a number of state-of-the-art DE variants, but also a number of other distributed DE and high-performing evolutionary algorithms. The speedup is significant especially on expensive problems, offering high potential in a broad range of real-world applications.

LanguageEnglish
Article number7530859
Pages704-716
Number of pages13
JournalIEEE Transactions on Parallel and Distributed Systems
Volume28
Issue number3
Early online date3 Aug 2016
DOIs
Publication statusPublished - 1 Mar 2017

Fingerprint

Cloud computing
Evolutionary algorithms
Costs
Virtual machine

Keywords

  • adaptive migration strategy
  • cloud computing
  • differential evolution
  • evolutionary algorithm
  • heterogeneous parallelism

Cite this

Zhan, Zhi-Hui ; Liu, Xiao-Fang ; Zhang, Huaxiang ; Yu, Zhengtao ; Weng, Jian ; Li, Yun ; Gu, Tianlong ; Zhang, Jun. / Cloudde : a heterogeneous differential evolution algorithm and its distributed cloud version. In: IEEE Transactions on Parallel and Distributed Systems. 2017 ; Vol. 28, No. 3. pp. 704-716.
@article{7a9b93ec1197466897f97bdca5b99695,
title = "Cloudde: a heterogeneous differential evolution algorithm and its distributed cloud version",
abstract = "Existing differential evolution (DE) algorithms often face two challenges. The first is that the optimization performance is significantly affected by the ad hoc configurations of operators and parameters for different problems. The second is the long runtime for real-world problems whose fitness evaluations are often expensive. Aiming at solving these two problems, this paper develops a novel double-layered heterogeneous DE algorithm and realizes it in cloud computing distributed environment. In the first layer, different populations with various parameters and/or operators run concurrently and adaptively migrate to deliver robust solutions by making the best use of performance differences among multiple populations. In the second layer, a set of cloud virtual machines run in parallel to evaluate fitness of corresponding populations, reducing computational costs as offered by cloud. Experimental results on a set of benchmark problems with different search requirements and a case study with expensive design evaluations have shown that the proposed algorithm offers generally improved performance and reduced computational time, compared with not only conventional and a number of state-of-the-art DE variants, but also a number of other distributed DE and high-performing evolutionary algorithms. The speedup is significant especially on expensive problems, offering high potential in a broad range of real-world applications.",
keywords = "adaptive migration strategy, cloud computing, differential evolution, evolutionary algorithm, heterogeneous parallelism",
author = "Zhi-Hui Zhan and Xiao-Fang Liu and Huaxiang Zhang and Zhengtao Yu and Jian Weng and Yun Li and Tianlong Gu and Jun Zhang",
year = "2017",
month = "3",
day = "1",
doi = "10.1109/TPDS.2016.2597826",
language = "English",
volume = "28",
pages = "704--716",
journal = "IEEE Transactions on Parallel and Distributed Systems",
issn = "1045-9219",
number = "3",

}

Cloudde : a heterogeneous differential evolution algorithm and its distributed cloud version. / Zhan, Zhi-Hui; Liu, Xiao-Fang; Zhang, Huaxiang; Yu, Zhengtao; Weng, Jian; Li, Yun; Gu, Tianlong; Zhang, Jun.

In: IEEE Transactions on Parallel and Distributed Systems, Vol. 28, No. 3, 7530859, 01.03.2017, p. 704-716.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Cloudde

T2 - IEEE Transactions on Parallel and Distributed Systems

AU - Zhan, Zhi-Hui

AU - Liu, Xiao-Fang

AU - Zhang, Huaxiang

AU - Yu, Zhengtao

AU - Weng, Jian

AU - Li, Yun

AU - Gu, Tianlong

AU - Zhang, Jun

PY - 2017/3/1

Y1 - 2017/3/1

N2 - Existing differential evolution (DE) algorithms often face two challenges. The first is that the optimization performance is significantly affected by the ad hoc configurations of operators and parameters for different problems. The second is the long runtime for real-world problems whose fitness evaluations are often expensive. Aiming at solving these two problems, this paper develops a novel double-layered heterogeneous DE algorithm and realizes it in cloud computing distributed environment. In the first layer, different populations with various parameters and/or operators run concurrently and adaptively migrate to deliver robust solutions by making the best use of performance differences among multiple populations. In the second layer, a set of cloud virtual machines run in parallel to evaluate fitness of corresponding populations, reducing computational costs as offered by cloud. Experimental results on a set of benchmark problems with different search requirements and a case study with expensive design evaluations have shown that the proposed algorithm offers generally improved performance and reduced computational time, compared with not only conventional and a number of state-of-the-art DE variants, but also a number of other distributed DE and high-performing evolutionary algorithms. The speedup is significant especially on expensive problems, offering high potential in a broad range of real-world applications.

AB - Existing differential evolution (DE) algorithms often face two challenges. The first is that the optimization performance is significantly affected by the ad hoc configurations of operators and parameters for different problems. The second is the long runtime for real-world problems whose fitness evaluations are often expensive. Aiming at solving these two problems, this paper develops a novel double-layered heterogeneous DE algorithm and realizes it in cloud computing distributed environment. In the first layer, different populations with various parameters and/or operators run concurrently and adaptively migrate to deliver robust solutions by making the best use of performance differences among multiple populations. In the second layer, a set of cloud virtual machines run in parallel to evaluate fitness of corresponding populations, reducing computational costs as offered by cloud. Experimental results on a set of benchmark problems with different search requirements and a case study with expensive design evaluations have shown that the proposed algorithm offers generally improved performance and reduced computational time, compared with not only conventional and a number of state-of-the-art DE variants, but also a number of other distributed DE and high-performing evolutionary algorithms. The speedup is significant especially on expensive problems, offering high potential in a broad range of real-world applications.

KW - adaptive migration strategy

KW - cloud computing

KW - differential evolution

KW - evolutionary algorithm

KW - heterogeneous parallelism

UR - http://www.scopus.com/inward/record.url?scp=85013053105&partnerID=8YFLogxK

UR - https://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=71

U2 - 10.1109/TPDS.2016.2597826

DO - 10.1109/TPDS.2016.2597826

M3 - Article

VL - 28

SP - 704

EP - 716

JO - IEEE Transactions on Parallel and Distributed Systems

JF - IEEE Transactions on Parallel and Distributed Systems

SN - 1045-9219

IS - 3

M1 - 7530859

ER -