Job Scheduling Optimization using BAT Algorithm in Hadoop Environment

The role of Hadoop in the big data environment is meant to be massive; it needs special attention for processing large data sets in a distributed environment. The role of HDFS is to store the data and Mapreduce is used for processing the data, where the data job is submitted by the master node and the task is assigned to slave node for effective processing. The aim of this article is to reduce the execution time and cost by improving the reliability. Here we use bio inspired algorithm known as BAT algorithm, it identifies the slave node accurately which helps the master node to assign task quickly. It also helps to find the sticky slots and avoid job starvation by reducing the scheduling complexity. The BAT algorithm produces better results by outperforming other basic randomized evolutionary algorithm.