VOLUMEPOOL: Identifying groups of volumes where data sets are recalled

Explanation: VOLUMEPOOL is an optional parameter specifying a set of volumes where a data set is recalled based on the volume from which the data set migrated. A data set that migrates from a volume in the volume pool is recalled to a volume in the volume pool.

VOLUMEPOOL(poolid): For poolid, substitute from 1 to 8 alphanumeric characters for the name of the volume pool.

Subparameter	Explanation
VOLUMES	VOLUMEPOOL(poolid VOLUMES(volser)) specifies the volumes belonging to the volume pool where the data sets are recalled. When you specify multiple volumes for a pool ID, DFSMShsm tries to recall data sets to the volume with the most space. If there is not enough space on the volume, DFSMShsm tries recalling the data set to up to four different volumes in the volume pool. For volser, substitute the serial number of the primary volume or volumes making up the volume pool where data sets are recalled.

Subparameter

Explanation

VOLUMES

VOLUMEPOOL(poolid VOLUMES(volser)) specifies the volumes belonging to the volume pool where the data sets are recalled. When you specify multiple volumes for a pool ID, DFSMShsm tries to recall data sets to the volume with the most space. If there is not enough space on the volume, DFSMShsm tries recalling the data set to up to four different volumes in the volume pool.

For volser, substitute the serial number of the primary volume or volumes making up the volume pool where data sets are recalled.

Defaults: None.

Note:

In a JES3 environment, at least one of the volumes in the volume pool must be added by the ADDVOL command. This JES3 requirement prevents the user from defining a volume pool that cannot be changed and fails the recall every time it is selected. It fails the recall because a volume cannot be selected for recall unless it has been added by the ADDVOL command. It cannot be changed, except at DFSMShsm startup, because the DEFINE and ADDVOL commands can be used only at DFSMShsm startup.
In a JES2 environment, no volume in the volume pool needs to be added with the ADDVOL command or mounted for the DEFINE command. The RECALL command, however, still requires a volume added by the ADDVOL command before the volume can be selected for the recall. After DFSMShsm startup, the ADDVOL command can be issued to add a volume, or the DEFINE command can be reissued with a new combination of volumes to change the association between volumes within a volume pool.
No checking is done to see if a volume is SMS-managed. A data set migrated from a volume that has been converted to SMS is recalled (if it is to be recalled as non-SMS) to a non-SMS-managed volume in the volume pool. The pool is not used for data sets recalled as SMS.
If poolid is specified without VOLUMES being specified, the volume pool becomes empty.
The hierarchy of selecting a candidate volume for recall is as follows:
- A limit of 140 volumes can be defined to a volume pool.
- If a data set being recalled is associated with a data set pool, the candidate volumes are selected from the data set pool.
- If the volume from which the data set migrated is part of a volume pool, the candidate volume is selected from the volume pool.
- If a volume is defined in multiple volume pools, the first definition is used. For example, if a data set was migrated from volume VP0001 and is recalled, it is recalled to volume pool SAMPLE1:
```
DEFINE VOLUMEPOOL(SAMPLE1 VOLUMES(VP0001 VP0003))
DEFINE VOLUMEPOOL(SAMPLE2 VOLUMES(VP0001 VP0002 VP0003))
```
- If the volume from which the data set migrated is not part of a volume pool, the candidate volumes are selected based upon the SETSYS RECALL values of ANYSTORAGEVOLUME or PRIVATEVOLUME.