Efficient MapReduce Framework Using Summation

Nowadays, data is the vital aspect of different activities. Thus, vast data is generated each and every second everywhere like industry, academics, health care, social networking, etc. As a result, an intelligent data analysis tool is needed. MapReduce (MR) framework is normally designed to process huge data and to support potential decision-making. MR has three phases—“map, shuffle, and reduce.” The intermediate data of mapper is sent to the reduce via shuffle phase. Thus, the massive traffic is generated within the shuffle phase. Although several efforts are done to raise the performance, the network traffic produced within the shuffle phase is generally ignored, which plays an essential role in performance improvement. This paper introduces a mechanism to increase the efficiency of MapReduce, which incorporate summation function at shuffle phase. It uses distributed election algorithm to incorporate the summation function. This proposed mechanism minimizes the network traffic (up to 55%) before sending to reduce function to incorporate summation which in turn enhance the overall performance of MR framework.