Naming convention for HALDB data definition names (ddnames)
IMS defines HALDB data definition names (ddnames) by appending a 1-byte suffix to the partition name. The suffix indicates the type of data set and, if you use multiple data set groups, differentiates the data sets within the group.
The following table shows the HALDB data set types and the corresponding ddname suffixes.
| Data set type | Ddname suffix | Additional suffixes if HALDB Online Reorganization is used |
|---|---|---|
| Database data set | A–J | M–V |
| Primary index (PHIDAM only) | X | Y |
| Indirect list data set (PHDAM and PHIDAM only) | L | L (the suffix for the ILDS does not change) |
If you use multiple data set groups, the A through J suffixes are the values that you would specify on the DSGROUP parameter in either a SEGM statement or, if support for DDL is enabled, a CREATE or ALTER TABLE DDL statement. The letter A identifies the first database data set (DBDS), the letter B identifies the second, and so forth, up to the letter J. If you do not use multiple data set groups, you do not specify the DSGROUP parameter and the ddname for the single data set that contains the record segments has the suffix A.
The suffixes M–V and Y are created automatically for the integrated HALDB Online Reorganization function of IMS. You do not need to specify them in the DBD. If you have never used the HALDB Online Reorganization function to reorganize a given partition, the suffixes M–V and Y are not used in that partition.
In PSINDEX databases, each partition contains only one data set. The suffix A is used for the ddname that corresponds to that data set.
For example, a PHIDAM database partition named PART1 would have ddnames of PART1A for its first DBDS, PART1B for the second DBDS, up to PART1J for the tenth DBDS. The indirect list data set (ILDS) and the primary index of partition PART1 would have ddnames of PART1L and PART1X, respectively. And a PSINDEX database partition named PARSI would have a ddname of PARSIA for its data set.
When reorganizing a partition, the integrated HALDB Online Reorganization function of IMS uses an additional data set for each data set that is active prior to starting the online reorganization process. For example, a ddname of PART1M is created to correspond to the active data set PART1A. A PART1N is created for PART1B, and so on, up to PART1V for PART1J, if it exists.
The ddnames must be unique among the database names, partition names, and Fast Path area names that are registered in the RECON data set.