Optimal scheduling workflows in cloud computing environment using Pareto‐based Grey Wolf Optimizer

Abstract

A workflow consists of dependent tasks, and scheduling of a workflow in a cloud environment means the arrangement of tasks of the workflow on virtual machines (VMs) of the cloud. By increasing VMs and the diversity of task size, we have a huge number of such arrangements.Finding an arrangement with minimum completion time among all of the arrangements is an Non-Polynomial-hard problem. Moreover, the problem becomes more complex when a scheduling should consider a couple of conflicting objectives. Therefore, the heuristic algorithms have been paid attention to figure out an optimal scheduling. This means that although the singleobjective optimization, ie, minimizing completion time, proposes the workflow scheduling as an NP-complete problem, multiobjective optimization for the scheduling problem is confronted with a more permutation space because an optimal trade-off between the conflicting objectives is needed. To this end, we extended a recent heuristic algorithm called Grey Wolf Optimizer (GWO) and considered dependency graph of workflow tasks. Our experiment was carried out using the WorkflowSim simulator, and the results were compared with those of 2 other heuristic task scheduling algorithms.to optimize some conflicting objectives stated previously in the cloud computing environment. We called it Pareto-based GWO (PGWO) and evaluated its performance on 2 patterns of workflows.Similar to the PSO algorithm, the GWO algorithm is based on swarm intelligence proposed by Mirjalili et al. 6 To evaluate the performance of the proposed algorithm practically, we implemented the extend algorithm using the WorkflowSim tool, 7 which is based on the CloudSim tool. Results of the proposed algorithm were compared with those of Strength Pareto Evolutionary Algorithm 2 (SPEA2) algorithm. 8The rest of the article is organized as follows. In Section 2, related work is presented in the greedy, single, and multiobjectives categories.In Section 3, first we introduce the concept of task scheduling and then we present our method in 2 parts, the extension of (1) the single-objective optimization to multiobjective optimization (MOO) and (2) the GWO algorithm for support of MOO. In Section 4, experimental results, simulation environment, performance indices, and results of the proposed algorithm are evaluated. Finally, conclusions are drawn in Section 5.

A PSO‐based task scheduling algorithm improved using a load‐balancing technique for the cloud computing environment

“…In addition to considering the interests of users, Khalili and Babamir considered the requirements of QoS for service providers. 24 They presented the gray wolf optimizer (GWO)-based algorithm as a multi-objective scheduling algorithm with a Pareto front with the aim of reducing makespan and cost, and increasing throughput. They compared their results with those of the strength Pareto evolutionary algorithm (SPEA2).…”

A Review of Cost and Makespan-Aware Workflow Scheduling in Clouds

“…This paper mainly focuses on the optimization of makespan and cost, but the QoS of users is also in°uenced by other factors, such as reliability and energy consumption. In this section, we divide the literatures [60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75][76][77][78][79] into three subsections based on the di®erent optimization objectives. They are brie°y introduced in Tables 9-11.…”

“…A few researches consider both the requirements of users and cloud providers. To address the con-°i cting objectives, Wu et al 73 and Khalili and Babamir 74 proposed respectively a market-oriented hierarchical work°ow scheduling strategy in clouds. Wu et al proposed a scheduling approach that mainly consists of two levels, namely servicelevel scheduling and task-level scheduling.…”

An efficient virtual machine allocation algorithm for parallel and distributed simulation applications

“…64 Such an optimization problem can be solved using heuristic algorithms such as genetic algorithm (GA), particle swarm optimization algorithm (PSOA), and ant colony optimization algorithm (ACOA). Meanwhile, how to select a certain number of VMs to make the running time of an application shortest is an NP-hard problem.…”

“…Meanwhile, how to select a certain number of VMs to make the running time of an application shortest is an NP-hard problem. 64 Such an optimization problem can be solved using heuristic algorithms such as genetic algorithm (GA), particle swarm optimization algorithm (PSOA), and ant colony optimization algorithm (ACOA). In this paper, GA is used to search the optimal combination of VMs.…”