Migration policies
Migration policies contain migration and storage media characteristics for data archived using IBM® Content Manager OnDemand. The information is used by the archived storage management process (ASM), also referred to as the migration process, to determine if and when archived data should be moved as it ages through a hierarchy of storage media having different performance and capacity characteristics. Examples of these archive media types include cloud storage, disk pools, optical, and tape storage. Each step in the movement of data through this storage hierarchy is referred to as a migration policy storage level, or simply, a storage level. Each policy must contain at least one storage level. Additional levels can be defined to meet your storage and retrieval requirements. The Start Archived Storage Management (STRASMOND) command is the command used to force data to move to the next storage level in the migration policy. The ASM process produces multiple reports that list the actions that the ASM process performed. The reports are produced with spooled file names beginning with QPRLCASM, such as QPRLCASM, QPRLCASMA, QPRLCASMD, and QPRLCASMU. The process also loads the same reports into the Content Manager OnDemand System Log, automatically starting the Content Manager OnDemand server for you if the server is not started so that the reports can be loaded. These reports should be checked each time ASM is run to ensure that processing of the data completed successfully. If you find a failure, you should check the job log for the STRASMOND job to determine the cause of the failure.
- If separate archived files are to be aggregated, or combined, with other archived files having similar retention and migration characteristics. Enabling aggregation is usually recommended because it can improve performance by allowing Content Manager OnDemand to manage a smaller number of larger objects rather than a large number of small objects. However, it is important to understand that an aggregate must reach its maximum size or exceed its time period before the aggregated object can flow to the first level of the Migration Policy. Because of this, you should be cautious of large aggregate sizes. If you are aggregating many small objects, it is possible for the aggregate not to be migrated to the first level of the Migration Policy for many months. For this reason, you might prefer to aggregate after a specified time period rather than by size.
- If two copies of archived data are to be kept at some or all levels in the migration sequence.
- If you choose tape or cloud storage for your archive storage media type, you can optionally choose to stage the data back to disk for a number of days after the data has been retrieved.
- If a one-time tape backup is performed.
The migration policy name can be up to 60 characters long and must not be a duplicate of another policy within the same instance. If the Enable aggregation is selected, the archived storage management process combines individual archived objects on disk into larger objects to provide efficient processing. This process occurs prior to migration of the object from disk to the first storage level. The aggregation process appends to the same file (aggregate) until the aggregate is 'closed.' The aggregate is closed when it either reaches a specified maximum size or a specified number of days. Storage levels can be added before or after a pre-existing storage level. Existing storage levels within a migration policy can also be changed or removed, however these changes will not affect migrated data already residing at this level. If you need to change the dates for migrated data already residing at a particular level, use the Change Policy Level Date (CHGPLDOND) command. If a one-time tape backup is requested, a tape media type must be specified.