Hive and MySQL - Query performance comparison for a multidimensional model

This paper presents a performance comparison of queries in multidimensional database stored in two different platforms, MySQL, which is one of the most popular database management systems for relational data, and Hive, which is a data warehouse and Big Data platform. We analyze the impact of the increase in the data volume on the query processing time on both platforms. The Brazilian Military Service data set was used with four data warehouse queries. Both Hive and MySQL environments were created in Google Cloud, using MySQL version 5.7 and two clusters for Hive, the first one with two workers, and the second one with four workers. The results showed that, as the volume of data grew, Hive's four workers cluster combined with Parquet file format had an equivalent or better performance than MySQL for the queries executed.