The relational database contains the cluster data, organised into tables, for reporting and analysis.
Data is stored and organized in tables within the database. The organization of this data is defined in the data schema of the tables.
The database and its data schema are partitioned for Platform Analytics data. A partitioned database has tables divided into multiple, smaller tables. This improves database performance for larger clusters.
In a large database, purging old job records, transforming data, and other database maintenance tasks can have a significant effect on database performance. Purging old job records and transforming data from smaller tables has less of an impact on the system performance of active tables than on larger tables.
The database tables are partitioned by quarter. Platform Analytics keeps three years of data in the database. Every month, Platform Analytics has a scheduled task that drops any partition that is older than three years by quarter.