Profiling-Based Effective Resource Utilization in Cloud Environment Using Divide and Conquer Method

Abstract

“The Cloud” in IT perception is simply a storage which is used to store data persistently through the resources that are available virtually. Computing on the cloud (cloud computing) means storing and accessing data over the Internet (the Cloud). The cloud computing is also regarded as a type of Internet-based computing. Effective resource utilization under cloud framework is one of the prime issues. Load balancer’s goal is to provide resources to serve incoming requests as well as fair resource allocation. Its job is to distribute the workload across the available computing resources. Auto-scaling basically provides flexibility with the number of available servers. Most of the load balancing algorithms suffer from starvation due to lack of utilization statistics. In this work, we incorporated profiling of resources in order to utilize them efficiently. Experimental results show that proposed algorithm outperforms traditional load balancing algorithm by 10–20%.