The effect of collocation on resources and system performance depends on the type of operation that is being run.
Table 1 summarizes the effects of collocation on operations.
| Operation | Collocation Enabled | Collocation Disabled |
|---|---|---|
| Backing up, archiving, or migrating client files | More media mounts to collocate files. | Fewer media mounts are required. |
| Restoring, retrieving, or recalling client files | Large numbers of files can be restored, retrieved, or recalled more quickly because files are on fewer volumes. | Multiple mounts of media might be
required for a single user because files might be spread across multiple
volumes. More than one user's files can be stored on the same sequential-access storage volume. For example, if two users try to recover a file that is on the same volume, the second user is forced to wait until the first user's files are recovered. |
| Storing data on tape | The server attempts to use all available tape volumes to separate user files before it uses all available space on every tape volume. | The server attempts to use all available space on each tape volume before you use another tape volume. |
| Media mount operations | More mount operations when user files
are backed up, archived, or migrated from client nodes directly to
sequential-access volumes. More mount operations during reclamation and storage pool migration. More volumes to manage because volumes are not fully used. |
More mount operations are required during restore, retrieve, and recall of client files. |
| Generating backup sets | Less time is spent searching database entries and fewer mount operations. | More time is spent searching database entries and fewer mount operations. |
When collocation is enabled for a group, single client node or file space, all the data that belongs to the group, the node, or the file space is moved or copied by one server process. For example, if data is collocated by group, all data for all nodes that belong to the same collocation group is migrated by the same process.
When collocating data, the Tivoli® Storage Manager server tries to keep files together on a minimal number of sequential-access storage volumes. However, when the server is backing up data to volumes in a sequential-access storage pool, the backup process has priority over collocation settings. As a result, the server completes the backup, but might not be able to collocate the data.
For example, suppose that you are collocating by node and you specify that a node can use two mount points on the server. Suppose also that the data that is backed up from the node can easily fit on one tape volume. During backup, the server might mount two tape volumes, and the node's data might be distributed across two tapes, rather than one. When you used collocation, the following server operations use one server process:
When migrating data from a random disk storage pool to a sequential storage pool, and collocation is by node or file space, nodes or file spaces are automatically selected for migration based on the amount of data to be migrated. The node or file space with the most data is migrated first. If collocation is by group, all nodes in the storage pool are evaluated to determine which node has the most data. The node with the most data is migrated first along with all the data for all the nodes that belong to that collocation group. This process takes place, regardless of how much data is stored in the file spaces of nodes and regardless of whether the low migration threshold was reached.
However, when migrating collocated data from a sequential storage pool to another sequential storage pool, the server orders the volumes according to the date when the volume was last accessed. The volume with the earliest access date is migrated first, and the volume with the latest access date is migrated last.
One reason to collocate by group is that individual client nodes often do not have sufficient data to fill high-capacity tape volumes. Collocating data by groups of nodes can reduce unused tape capacity by putting more collocated data on individual tapes. Also, collocating data by groups of file spaces reduces the unused tape to a finer degree.
The data that belongs to all the nodes in the same collocation group are migrated by the same process. Therefore, collocation by group can reduce the number of times a volume that is going to be migrated needs to be mounted. Collocation by group can also minimize database scanning and reduce tape passes during data transfer from one sequential-access storage pool to another.