Linear Regression Assisted Prediction Based Load Balancer For Cloud Computing

Abstract

Cloud Computing is one of the most ubiquitous technologies today. Cloud service providers are required to provide services to its users productively and efficiently. Resources must be selected and allocated properly based on the task’s attributes. The objective of a load balancing algorithm is to ensure efficient and fair distribution of load among all the computing resources. The challenge today, however, is to maintain performance standards in spite of the rapidly increasing data and performance needs of the users. This motivates an automated approach towards load balancing. In this work, we proposed a load balancing algorithm based on the usage statistics which predicts the queue of virtual machines for the optimal usage of cloud resources. Experimental results shows the improvement in response time with compared to the conventional approach.