Multi-Objective Task and Workflow Scheduling Approaches in Cloud Computing: a Comprehensive Review
- PDF / 3,263,545 Bytes
- 30 Pages / 547.087 x 737.008 pts Page_size
- 35 Downloads / 298 Views
Multi-Objective Task and Workflow Scheduling Approaches in Cloud Computing: a Comprehensive Review Mehdi Hosseinzadeh & Marwan Yassin Ghafour & Hawkar Kamaran Hama & Bay Vo & Afsane Khoshnevis
Received: 15 April 2019 / Accepted: 20 August 2020 # Springer Nature B.V. 2020
M. Hosseinzadeh Institute of Research and Development, Duy Tan University, Da Nang 550000, Vietnam e-mail: [email protected]
goals which have been utilized by various schemes to solve the task and workflow scheduling problems. This paper focuses on the metaheuristic multi-objective optimization context and presents a comprehensive survey and overview of the multi-objective scheduling approaches designed for various cloud computing environments. It classifies the scheduling schemes regarding their applied multi-objective optimization algorithms and describes how they have adapted the optimization algorithms to solve scheduling problems. Furthermore, a comparison of the multi-objective scheduling schemes is provided, which illuminates future research directions, and finally concluding remarks are presented.
M. Hosseinzadeh Health Management and Economics Research Centre, Iran University of Medical Sciences, Tehran, Iran
Keywords Cloud . Task . Workflow . Scheduling . Energy . Optimization . NSGA-II . MOEA . Pareto front
M. Y. Ghafour Department of Computer Science, College of Science, University of Halabja, Halabja, Iraq e-mail: [email protected]
1 Introduction
Abstract Efficient task and workflow scheduling are very important for improving resource management and reducing power consumption in cloud computing data centers (DCs). However, regarding numerous tasks, virtual machines, and several objectives which should be taken into account, scheduling is considered to be an NP-Hard problem. Multi-objective optimization is an interesting technique to deal with multiple conflicting
H. K. Hama Computer Science Department, College of Basic Education, University of Sulaimani, Sulaymaniyah, Iraq e-mail: [email protected] B. Vo (*) Faculty of Information Technology, Ho Chi Minh City University of Technology (HUTECH), Ho Chi Minh City, Vietnam e-mail: [email protected] A. Khoshnevis Department of Computer Engineering, Urmia Branch, Islamic Azad University, Urmia, Iran e-mail: [email protected]
Nowadays, cloud computing has become a crucial part of information technology, aimed to provide required virtual resources for back-end processing in various multi-tier applications. Infrastructure-as-a-Service clouds, such as Amazon EC2, introduce multiple kinds of virtual machines (VMs) [1], which can have different amounts of various resources. This allows customers to run their programs on the most appropriate VMs to achieve their required level of Quality of Service (QoS). Accordingly, Cloud Service Providers (CSPs) can charge customers by a pay-as-you-go model regarding the specification of their provided VMs [2]. In
M. Hosseinzadeh et al.
addition, cloud computing can provide access to various software, and its DCs
Data Loading...
