Over the last few blogs, we have stepped through a project to implement Information Lifecycle Management (ILM) on corporate databases. First, we evaluated the target databases, then we determined the Business Objects and assigned retention periods to the data – including both Legal and Operational requirements. Now that we are ready to start applying the retention policies and deleting data, it is a good idea to set up an archive database as an intermediate repository for business objects classified as legal records.
Rather than just deleting data out of the source, you should create an archive database that resides on lower cost infrastructure. Create the archive table definitions identical to source and test the SQL statements defined in the previous step (classification). A database archiving tool like Informatica’s Data Archive integrates this step into the product making it much easier to setup and manage. Once data becomes read-only and is less frequently accessed, it should be moved to the archive until the retention period expires. A convenient view for the source application to see the moved data should then be created. Using this view, business users have all the access they need to the aged data.
By moving a bulk of the information out of the production database, the application’s performance will improve drastically. Additionally, all copies made for test, development and backup will be that much smaller reducing infrastructure needs and cost.
With the aged data residing in an archive database, you can then execute the data destruction process and start purging data in a central environment. If you can consolidate all of your database records to a single archive, searching for records during an eDiscovery is easier because all of the structured content resides in one place. For operational data that has shorter retention periods, this information should be cleaned out of the source database on a regular, frequent schedule.
Before you can claim victory, there are a few cleanup steps that need to occur before you can realize the true benefit of deleting database data. In our next and final blog of this series, we will review the final necessary tasks and then calculate our benefit.
Julie Lockner, President & Founder, CentricInfo
