Monitoring
Use the tools in your relational database management system (RDBMS) to monitor the growth of the data warehouse, the system resources utilized by the data warehouse, and the queries issued against the data warehouse. For example, you can use the statement explain utilities to analyze the queries, understand the access plans, and make adjustments. To help you identify bottlenecks and resource utilization on the database, you can monitor memory, CPU I/O activity, and data growth on the database. Make use of your RDBMS tools to monitor the database.
Maintaining
The sample database model contains the routines to compute statistics for DB2® and Oracle Server at certain stages of the ETL (extract, transform, and load) process. If you customize the data warehouse model by adding tables to it, you might want to make sure that the statistics of these tables are also updated. Over time, as the database grows, the collecting statistics and reorganizing data might become time-consuming. You might want to stop gathering statistics and start sampling or you might want to change the frequency of statistics collection by including only those tables that have changed during the ETL process. For more information, see the documentation of your relational database management system.
Taking back ups
Set up a plan for disaster recovery. The events (DDL or ETL changes), frequency, and type of backup depends on factors such as downtime allowed for the database for the backup activity, size of the database, time required for the backup activity, and the amount of data loss tolerated in the event of a disaster. Your relational database management system (RDBMS) has utilities for backing up databases, as well as backup strategies such as full, off-line, on-line, incremental and delta. For more information, see your RDBMS documentation.