Biased Random Sampling with Firefly Optimization (BRS-FO) based on Load Balancing for Virtual Machine Migration in Cloud Computing.

Abstract

A concept known as "Cloud Computing" aims to simplify the on-demand delivery of software, hardware, as well as data as services and give end users adaptable, scalable, and accessible services through the Internet. The major goal of the suggested technique is to create a healthy balance of load across all the Cloud resources servers while maximizing resource usage. Every resource will first have a load model created based on a number of variables, including memory use, processing time, and access rate. Several meta heuristics optimization algorithm are presented in literature for VM migration with load balancing in Cloud Computing (CC). However, in the paper Biased Random Sampling with Firefly Optimization (BRS-FO) was combined. The Load balancing is performed by biased random sampling and Firefly Optimization by maintaining the virtual server availability. In this method the performance of proposed algorithm was compared with PSO, GA and Honey Bee Optimization (HBO). The parameters taken for analysis are Makespan, Response time and energy consumption. From this experimental results, the proposed BRS-FO achieved the makespan of 5s, response time of 1s and energy consumption of 5J and proved this method is efficient than other system