Reorganizing a database and using PSSR to sort the unloaded
data set
You can include a PSSR job step in your reorganization
job to sort the unloaded data set during reorganization process.
Before you begin
Several considerations apply for using the output of PSSR
as an input of IMS™ High
Performance Load for z/OS® Version
2 Release 1 or later. For details, see Considerations before using PSSR.
Procedure
To use PSSR, you must run several programs. These
programs can be run in either a single job or in several jobs. A typical
PSSR job stream contains some or all of the following steps:
Recommendation: Always include steps 1 and 13 in the PSSR job that
actually performs a database reorganization. Use of any of the other
steps depends on the logical relationships involved.
Note: For
information about using IMS utilities,
see IMS Database
Utilities.
Run the IMS Database Image Copy utility
(DFSUDMP0) to create a copy of each of the old databases before reorganization.
These image copies are the backup copies of your databases.
These backup copies can be used if something goes wrong in the reorganization
process.
Run the IMS Database
Prereorganization utility (DFSURPR0) to create a control data set
(DFSURCDS).
This control data set is used by the other IMS logical relationship resolution
utilities. The DFSURPR0 utility also indicates which databases and
segments (if any) need to be scanned by the IMS Database Scan utility.
Run the IMS Database Scan utility
(DFSURGS0) to scan any databases not being loaded or reorganized for
segments that contain logical relationships that are affected by loading
or reorganizing other databases.
For each segment of
this type, the utility generates one or more output records, depending
upon the relationships in which that segment is involved. It generates
an output work data set that will be used by the IMS Database Prefix Resolution utility.
Run the IMS HD
Reorganization Unload utility
(DFSURGU0) or its equivalent program such as unload utilities of IMS High Performance Unload to create
a sequential data set that contains the old unloaded databases.
Run the DBDGEN procedure.
This IMS procedure creates the DBDs for the new reorganized
databases. You must assemble and link-edit the new DBDs.
Run the PSSR utility.
When
PSSR is run in either an IMS environment
or a non IMS environment, it
sorts the unloaded database data set in physical sequence.
Note: The
only difference between a standard database reorganization job and
a database reorganization job that is done by PSSR is the addition
of this step.
Running a PSSR job includes the following tasks:
Select the IBM® supplied
cataloged procedure to use, or prepare a similar procedure of your
own.
There are two ways to run PSSR, and the following
two catalog procedures are provided:
FABSPSSR
Use the FABSPSSR procedure if you want to run the PSSR job as
a standard (non IMS) MVS™ batch job.
FABSPSSI
Use the FABSPSSI procedure if you want to run the PSSR job as
a batch job using the IMS environment.
Interpret the output reports to verify that the process
completed successfully.
Run the IDCAMS program to delete the old
database data sets and allocate the new database data sets.
Run the IMS HD
Reorganization Reload utility
(DFSURGL0) or an equivalent program such as IMS HP Load to reload a database from the sequential
data set that is created by the IMS HD
Reorganization Unload utility
or its equivalent utility.
This step also creates work
data sets that are used as input to the logical relationship resolution
utilities.
Run the IMS Database Prefix Resolution utility
(DFSURG10) to accumulate the information that is generated in work
data sets (DFSURCDS and DFSURWF1).
It produces an output
data set (DFSURWF3) that contains the prefix information that is needed
to complete the logical relationships defined for the databases. It
also produces an output data set (DFSURIDX) that contains the information
that is needed to create secondary index databases.
Run the IMS HISAM Reorganization Unload utility
(DFSURUL0) to format the index work data set (DFSURIDX) to a form
that can be used by the IMS HISAM
Reorganization Reload utility.
Run the IMS HISAM Reorganization Reload utility
(DFSURRL0) to create the secondary index databases from the reorganized
output data sets provided by the IMS HISAM
Reorganization Unload utility.
Run the IMS Database Prefix Update utility
(DFSURGP0) to update the prefix of each segment of which prefix information
is affected by the reload process.
The utility uses
the output (DFSURWF3) from the IMS Database Prefix Resolution utility
to update the prefix of each segment.
Run the IMS Database Image Copy utility
(DFSUDMP0) to create a copy of each of the new databases.
These
copies are the first backup copies of the new databases.
FABSPSSR and FABSPSSI JCL procedures
The
following two figures show the contents of the FABSPSSR JCL procedure
and the FABSPSSI JCL procedure.