کنفرانس بین المللی محاسبات نرم

صفحه اصلی / چهارمین کنفرانس بین المللی محاسبات نرم

A Multi-objective PSO based method for Task Scheduling in Cloud Computing

نویسندگان :

Ali Bazghandi¹ Mostafa Bazghandi²

1- دانشگاه صنعتی شاهرود 2- دانشگاه آزاد اسلامی واحد سبزوار

کلمات کلیدی :

Scheduling, PSO, Migration, Trade-off, Parameters

چکیده :

Cloud computing is a technology in which shared resources (networks, servers, storage locations, applications and services) are available through an extensive and dynamic network, and the cloud user can get access to it as quickly as possible and at the lowest cost. In this paper, a scheduling method based on particle swarm optimization (PSO) is presented so that it can be used to establish a trade-off between various quality of service parameters while optimally allocating the resources and properly making a load balance. The proposed technique prevents overloading and under-loading servers by optimally assigning tasks to processing servers. Also, servers that are overloaded and in other words congested, their tasks are transferred to another servers by live migration of the virtual machines. It provides increasing in cloud infrastructure load balancing and thus the execution time of tasks are significantly reduced. The proposed method was compared with two Ant Colony Optimization (ACO) and Round Robin (RR) algorithms and it is shown that the proposed method is able to execute the desired requests in less time versus two algorithms. It is resulted by using the multi-objective PSO algorithm and the correct migrations that the proposed technique is provided. By using this method, it is made a trade-off among various quality of service parameters involved response time, energy consumption and degradation of the service level agreement.