Multi-objective Prediction based Task Scheduling Method in Cloud Computing

Abstract

Cloud Computing is Internet based computing where one can store and access their personal resources from any computer through Internet. Cloud Computing is a simple pay-per-utilize consumer-provider service model. Cloud is nothing but large pool of easily accessible and usable virtual resources. Task (Job) scheduling is always a noteworthy issue in any computing paradigm. Due to the availability of finite resources and time variant nature of incoming tasks it is very challenging to schedule a new task accurately and assign requested resources to cloud user. Traditional task scheduling techniques are improper for cloud computing as cloud computing is based on virtualization technology with disseminated nature. Cloud computing brings in new challenges for task scheduling due to heterogeneity in hardware capabilities, on-demand service model, pay-per-utilize model and guarantee to meet Quality of Service (QoS). This has motivated us to generate multi-objective methods for task scheduling. In this research paper we have presented multi-objective prediction based task scheduling method in cloud computing to improve load balancing in order to satisfy cloud consumers dynamically changing needs and also to benefit cloud providers for effective resource management. Basically our method gives low probability value for not capable and overloaded nodes. To achieve the same we have used sigmoid function and Euclidean distance. Our major goal is to predict optimal node for task scheduling which satisfies objectives like resource utilization and load balancing with accuracy.