Big Data Solutions Proposed for Cluster Computing Systems Challenges: A survey

Abstract

CCS (Cluster Computing System) is coming to solve the problems of standard technology. Whose, objective is to improve the performance/power efficiency of a single processor for storing and mining the large data sets, using the parallel programming to read and process the massive data sets on multiple disks and CPUs. The thing which makes these systems somewhat performant than the standard technology is the physical organization of computing nodes in the cluster. Currently, this kind of cluster does not entirely solve the problem because it comes with its challenges, which are Node failures, Computations, Network Bottleneck, and Distributed programming. All these problems are coming when we are mining and storing the massive volume of data using cluster computing. To solve these challenges, Google invented a new Big Data framework of data processing called MapReduce, to manage large scale data processing across large clusters of commodity servers. The paper outlines the running of CCS and presents its challenges in this era of Big Data. Moreover, it introduces the most popular Big Data solutions proposed to overcome the CCS challenges. Also, it shows how Big Data technologies solve CCS issues. Generally, the main goal of this work is to provide a better understanding of the challenges of CCS and identify the essential big data solutions in this increasingly important area.