Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach

One of the most difficult aspects of scheduling operations on virtual machines in a multi-cloud environment is determining a near-optimal permutation. This task requires assigning various computing jobs with competing objectives to a collection of virtual machines. A significant number of NP-hard pr...

Full description

Saved in:
Bibliographic Details
Main Authors: Farid, Mazen, Heng, Siong Lim, Chin, Poo Lee, Latip, Rohaya
Format: Article
Published: Multidisciplinary Digital Publishing Institute 2023
Online Access:http://psasir.upm.edu.my/id/eprint/110241/
https://www.mdpi.com/2073-8994/15/11/2047
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.upm.eprints.110241
record_format eprints
spelling my.upm.eprints.1102412024-09-04T04:08:39Z http://psasir.upm.edu.my/id/eprint/110241/ Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach Farid, Mazen Heng, Siong Lim Chin, Poo Lee Latip, Rohaya One of the most difficult aspects of scheduling operations on virtual machines in a multi-cloud environment is determining a near-optimal permutation. This task requires assigning various computing jobs with competing objectives to a collection of virtual machines. A significant number of NP-hard problem optimization methods employ multi-objective algorithms. As a result, one of the most successful criteria for discovering the best Pareto solutions is Pareto dominance. In this study, the Pareto front is calculated using a novel multi-objective minimum weight approach. In particular, we use particle swarm optimization (PSO) to expand the FR-MOS multi-objective scheduling algorithm by using fuzzy resource management to maximize variety and obtain optimal Pareto convergence. The competing objectives include reliability, cost, utilization of resources, risk probability, and time makespan. Most of the previous studies provide numerous symmetry or equivalent solutions as trade-offs for different objectives, and selecting the optimum solution remains an issue. We propose a novel decision-making strategy named minimum weight optimization (MWO). Multi-objective algorithms use this method to select a set of permutations that provide the best trade-off between competing objectives. MWO is a suitable choice for attaining all optimal solutions, where both the needs of consumers and the interests of service providers are taken into consideration. (MWO) aims to find the best solution by comparing alternative weights, narrowing the search for an optimal solution through iterative refinement. We compare our proposed method to five distinct decision-making procedures using common scientific workflows with competing objectives: Pareto dominance, multi-criteria decision-making (MCDM), linear normalization I, linear normalization II, and weighted aggregated sum product assessment (WASPAS). MWO outperforms these strategies according to the results of this study. Multidisciplinary Digital Publishing Institute 2023 Article PeerReviewed Farid, Mazen and Heng, Siong Lim and Chin, Poo Lee and Latip, Rohaya (2023) Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach. Symmetry, 15 (11). art. no. 2047. pp. 1-28. ISSN 2073-8994 https://www.mdpi.com/2073-8994/15/11/2047 10.3390/sym15112047
institution Universiti Putra Malaysia
building UPM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Putra Malaysia
content_source UPM Institutional Repository
url_provider http://psasir.upm.edu.my/
description One of the most difficult aspects of scheduling operations on virtual machines in a multi-cloud environment is determining a near-optimal permutation. This task requires assigning various computing jobs with competing objectives to a collection of virtual machines. A significant number of NP-hard problem optimization methods employ multi-objective algorithms. As a result, one of the most successful criteria for discovering the best Pareto solutions is Pareto dominance. In this study, the Pareto front is calculated using a novel multi-objective minimum weight approach. In particular, we use particle swarm optimization (PSO) to expand the FR-MOS multi-objective scheduling algorithm by using fuzzy resource management to maximize variety and obtain optimal Pareto convergence. The competing objectives include reliability, cost, utilization of resources, risk probability, and time makespan. Most of the previous studies provide numerous symmetry or equivalent solutions as trade-offs for different objectives, and selecting the optimum solution remains an issue. We propose a novel decision-making strategy named minimum weight optimization (MWO). Multi-objective algorithms use this method to select a set of permutations that provide the best trade-off between competing objectives. MWO is a suitable choice for attaining all optimal solutions, where both the needs of consumers and the interests of service providers are taken into consideration. (MWO) aims to find the best solution by comparing alternative weights, narrowing the search for an optimal solution through iterative refinement. We compare our proposed method to five distinct decision-making procedures using common scientific workflows with competing objectives: Pareto dominance, multi-criteria decision-making (MCDM), linear normalization I, linear normalization II, and weighted aggregated sum product assessment (WASPAS). MWO outperforms these strategies according to the results of this study.
format Article
author Farid, Mazen
Heng, Siong Lim
Chin, Poo Lee
Latip, Rohaya
spellingShingle Farid, Mazen
Heng, Siong Lim
Chin, Poo Lee
Latip, Rohaya
Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
author_facet Farid, Mazen
Heng, Siong Lim
Chin, Poo Lee
Latip, Rohaya
author_sort Farid, Mazen
title Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
title_short Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
title_full Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
title_fullStr Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
title_full_unstemmed Scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
title_sort scheduling scientific workflow in multi-cloud: a multi-objective minimum weight optimization decision-making approach
publisher Multidisciplinary Digital Publishing Institute
publishDate 2023
url http://psasir.upm.edu.my/id/eprint/110241/
https://www.mdpi.com/2073-8994/15/11/2047
_version_ 1811686061812744192
score 13.209306