On the performance of SQL scalable systems on Kubernetes: a comparative study

Abstract The popularization of Hadoop as the the-facto standard platform for data analytics in context Big Data applications has led to upsurge SQL-on-Hadoop systems, which provide scalable query execution engines allowing use SQL queries on stored HDFS. In this context, Kubernetes appears leading choice simplify deployment and scaling containerized applications; however, there is a lack studies about performance systems deployed Kubernetes, gap we intend fill paper. We present an experimental study involving four representative platforms: Apache Drill, Hive, Spark Trino. Concretely, analyze these when they are cluster with by using TPC-H benchmark. results our can help practitioners users what expect terms if plan advantages deploy analyzed platforms.