A hybrid multi-objective metaheuristic optimization algorithm for scientific workflow scheduling
- PDF / 2,823,954 Bytes
- 25 Pages / 595.276 x 790.866 pts Page_size
- 13 Downloads / 169 Views
(0123456789().,-volV)(0123456789().,-volV)
A hybrid multi-objective metaheuristic optimization algorithm for scientific workflow scheduling Ali Mohammadzadeh1 • Mohammad Masdari1 • Farhad Soleimanian Gharehchopogh1 • Ahmad Jafarian2 Received: 29 May 2020 / Revised: 8 August 2020 / Accepted: 30 October 2020 Ó Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Workflow is composed of some interdependent tasks and workflow scheduling in the cloud environment that refers to sorting the workflow tasks on virtual machines on the cloud platform. We will encounter many sorting modes with an increase in virtual machines and the variety in task size. Reaching an order with the least makespan is an NP-hard problem. The hardness of this problem increases even more with several contradictory goals. Hence, a meta-heuristic algorithm is what required in reaching the optimal response. Thus, the algorithm is a hybridization of the ant lion optimizer (ALO) algorithm with a Sine Cosine Algorithm (SCA) algorithm and used it multi-objectively to solve the problem of scheduling scientific workflows. The novelty of the proposed algorithm was to enhance search performance by making algorithms greedy and using random numbers according to Chaos Theory on the green cloud computing environment. The purpose was to minimize the makespan and cost of performing tasks, to reduce energy consumption to have a green cloud environment, and to increase throughput. WorkflowSim simulator was used for implementation, and the results were compared with the SPEA2 workflow scheduling algorithm. The results show a decrease in the energy consumed and makespan. Keywords Ant lion optimizer Sine cosine optimization Meta-heuristic Green cloud computing Workflow scheduling
1 Introduction Cloud computing is one of the most known computing models that allow the user to store and process applications on large data centers. It allows users to access data remotely as well. Moreover, it enables organizations and companies to use high computing capacity without a high cost to build massive data centers. Overall, the cloud-computing & Mohammad Masdari [email protected] Ali Mohammadzadeh [email protected] Farhad Soleimanian Gharehchopogh [email protected] Ahmad Jafarian [email protected] 1
Department of Computer Engineering, Urmia Branch, Islamic Azad University, Urmia, Iran
2
Department of Mathematics, Urmia Branch, Islamic Azad University, Urmia, Iran
platform saves time and money [1]. For instance, providers like Amazon EC2 allow their users to access and manage virtual machines operating within the data centers. Although cloud computing has many benefits, managing cloud resources is critical, especially when several users present their applications simultaneously over the cloud. Hence, cloud computing needs an efficient scheduling strategy for resource management. Process scheduling is task mapping on available resources of the system without violating the limitations. A suitable schedule is designed
Data Loading...
