Performance Tuning Of Apache Spark Framework In Big Data Processing with Respect To Block Size And Replication Factor

Apache Spark has recently become the most popular big data analytics framework. Default configurations are provided by Spark. HDFS stands for Hadoop Distributed File System. It means large files will be physically stored on multiple nodes in a distributed fashion. The block size determines how distributed, while replication factor reliable are. If there is just one copy of each given file and node fails, unreadable. configurable per file. results analysis experimental study to determine efficiency adjusting settings tuning minimizing application execution time as compared standard values described this paper. Based vast number studies, we employed trial-anderror strategy fine-tune these values. We chose two workloads test framework comparative analysis: Wordcount Terasort. used elapsed evaluate same.