Process Optimization of Big-Data Cloud Centre Using Nature Inspired Firefly Algorithm and K-Means Clustering

Abstract

During the last decade, the growth of big data is immeasurable in information technology. Big data has the potential to take all the decisions necessary for a company or business. But it has many challenges as well. As its size and volume are immeasurably ample it is a very challenging task to store, process and mines it. At the same time as a boon to it cloud computing has a large capacity to store this big data and provides tremendous processing power. It is a challenging task to process large amount of data frequently in the big-data cloud center through the thousands of interconnected servers. Due to the day by day growth of the big-data, big-data cloud center is forced to improve its Quality of Service (QoS) metrics like throughput, latency and response time. Hence, to develop an optimal data processing optimization method is a current research problem that has to be solved. The major intention of this paper is to develop an application that provides maximum throughput, minimum latency and reduce the response time. Toward this, we have developed an optimization technique using nature-inspired firefly optimization algorithm and k-means clustering (FA-KMeans). The developed optimization method has been evaluated with state of art algorithms. Its experimental result elucidates that our proposed method provides good throughput, reduces latency and response time.