A Comparison of Query Execution Speeds for Large Amounts of Data Using Various DBMS Engines Executing on Selected RAM and CPU Configurations

In modern economies, most important business decisions are based on detailed analysis of available data. In order to obtain a rapid response from analytical tools, data should be pre-aggregated over dimensions that are of most interest to each business. Sometimes however, important decisions may require analysis of business data over seemingly less important dimensions which have not been pre-aggregated during the ETL process. On these occasions, the ad-hoc "online" aggregation is performed whose execution time is dependent on the overall DBMS performance. This paper describes how the performance of several commercial and non-commercial DBMSs was tested by running queries designed for data analysis using "ad-hoc" aggregations over large volumes of data. Each DBMS was installed on a separate virtual machine and was run on several computers, and two amounts of RAM memory were allocated for each test. Measurements of query execution times were recorded which demonstrated that, as expected, column-oriented databases out-performed classical row-oriented database systems.