International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 01 Issue: 04 | Apr 2023
p-ISSN: 2395-0072
www.irjet.net
Optimizing Task Scheduling in Mobile Cloud Computing Using Particle Swarm Optimization (PSO) Algorithm Shouaib Scander1, Souheil Khawatmi2, Yaser Fawaz 3 1pursuing a M. S. System and Computer Networks at the Faculty of Informatics Engineering,
University of Aleppo, Syria.
2associate professor at the Faculty of Informatics Engineering, University of Aleppo, Syria.
3associate professor at the Faculty of Informatics Engineering, University of Aleppo, Syria.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Cloud computing is a modern type of shared
the execution time and power consumption of any task run on mobile devices [5].
infrastructure that could interconnect large groups of systems and allows user to connect via the Internet. The term cloud is an expression used to refer to the internet and The Internet is the basis on which cloud computing depends. One of the most necessary requirements in cloud computing system is task scheduling, which plays a main role in the performance of each part of cloud computing equipment. Task scheduling determines tasks that need to be sent to the appropriate virtual device to meet user-defined quality of service (QoS) constraints such as completion time and cost in cloud. In our paper, we have proposed a comprehensive multi-purpose task scheduling optimization model which reduces task transmitting time, execution time and cost. The proposed model is built based on Particle Swarm Optimization (PSO), and the implementation results offer that the new proposed model is more dynamic in speeding up tasks execution and decreasing costs.
Although cloud computing provides many services, there are many problems with it. one of these problems, scheduling tasks that is one of the most critical problems due to the need to establish a suitable sequence to divide these tasks [6]. Therefore, cloud computing uses scheduling algorithms to assign confirmed tasks to nominated resources at an accurate time [7], mostly focused on cloud performance optimization which is bandwidth, memory and time discount. Tasks scheduling are split into two parts: one is used as a unified scheduler for the resource gathering, fundamentally responsible for scheduling cloud APIs and applications, and the other one is for scheduling unified port resources in the cloud such as task scheduler. [8].In this field, some researchers have published papers on the problem of scheduling tasks to get better performance using most optimization methods. The aim of these papers was to reduce cost, response time, uptime and resource usage [9].
Key Words: Offloading, Mobile Cloud Computing, Tasks Scheduling, Particle Swarm Optimization (PSO)
1.INTRODUCTION
The rest of this paper is organized as follows: section II presents Research motivation. section III presents Background and related works. section IV presents Mathematical models. section V presents Implementation and performance evaluation. section VI Conclusion and future work.
Cloud computing is a new type of computing that provides applications, data and all computer services dynamically over the internet [1]. Now it has become one of the most important fields of information technology. Organizations that need additional resources to develop their data centers rent these resources from the cloud computing system instead of purchasing those resources and pay according to their use of these additional resources [2]. Many applications require powerful computing devices and consume too much energy. Therefore, it is not a good idea to run such applications on constrained-resources devices, such as mobile devices, since they have limited computing power and battery life [3]. To handle this situation, researchers proposed tasks offloading to run the resourcehungry applications on the cloud [4]. Offloading is a technology by which large applications or parts of them i. e., tasks on local devices, are sent to the cloud to run there, as shown in Figure 1. Then, the results are sent back to the end devices, thus, the offloading process reduces
© 2023, IRJET
|
Impact Factor value: 8.226
Figure 1: Task scheduling-based cloud scheme
|
ISO 9001:2008 Certified Journal
|
Page 864