Auto Tuning of Hadoop and Spark parameters

Data of the order terabytes, petabytes, or beyond is known as Big Data. This data cannot be processed using traditional database software, and hence there comes need for Platforms. By combining capabilities features various big applications utilities, Platforms form a single solution. It platform that helps to develop, deploy manage environment. Hadoop Spark are two open-source provided by Apache. Both these platforms have many configurational parameters, which can unforeseen effects on execution time, accuracy, etc. Manual tuning parameters tiresome, automatic ways should needed tune them. After studying analyzing previous works in automating this paper proposes algorithms - Grid Search with Finer Tuning Controlled Random Search. The performance indicator studied Execution Time. These help automatically. Experimental results shown reduction time about 70% 50% 81.19% 77.77% Search, respectively.