Archiving Data Warehouse Data

This topic outlines steps that are needed to archive and delete unneeded Data Warehouse data.

See Deleting Data from the Data Warehouse for more information about how to properly delete Data Warehouse data.

To archive Data Warehouse data

  1. Make sure that there are no imports or report queries running against the Data Warehouse for the best archive performance.

  2. Copy the full production database to an archive database.

  3. Delete all the data from the archive database that is newer than what you want to remain on the archive database.

  4. Archive the database that has been reduced.

  5. Then, copy the full production database to another database (the new production database).

  6. Delete the data that was archived in steps 3 through 4 from the new production database.

  7. As soon as the old data has been removed from the 'new' production database, and the data has been verified, the 'new' production database can replace the current production database.

Some of the largest data tables (for example log user, URI) will not be affected by the deletes because the data in these tables is not date based. Insert performance after archiving may not change significantly because of the time that is needed to search these large tables.

See Also

Other Resources

Developing with the Data Warehouse Analytics System