

3. Installation › 3.5 Generate a Database Unit › 3.5.1 Define Catalog Indexes for Prefix and Tapeprefix
3.5.1 Define Catalog Indexes for Prefix and Tapeprefix
A database unit requires two data set name prefixes, a DASD
prefix (prefix) and a tape prefix (tapeprefix), which are
defined in 'prefix.MICS.PARMS(JCLDEF)', described in section
2.3.3.2.1, "JCL Option Definition (JCLDEF)". These data set
name prefixes serve several purposes:
1. They are used to give unique names to the data sets
associated with a specific database unit.
2. The high level qualifier of the prefixes may serve as a
catalog index for the data sets. This index may provide
a pointer to a user catalog.
3. The prefixes provide meaningful qualification for
identification and control of the CA MICS shared data
sets.
The prefix and tapeprefix for a database unit must be
different from the sharedprefix and from the prefixes and
tapeprefixes for all other database units in the database
complex. However, the prefix and tapeprefix for a unit may
be the same. Also, the sharedprefix and the other prefixes
may have a common high-level index.
Unless standards at your site prevent it, this step should be
used to ensure that the indexes established for the
tapeprefix and prefix point to a catalog other than the
master catalog. We strongly recommend that the master
catalog not be used for cataloging CA MICS data sets,
especially in sites where there may be more than one OS
system on which CA MICS may be used.
In selecting a high-level index for CA MICS data sets, be
aware of the following:
1. Most sites have data set naming standards for production
data sets. CA MICS should be considered as a production
system, and the high-level index that is chosen accordingly.
2. If a high-level index of a TSO userid is selected, ensure
that your site does not delete data sets that remain
unreferenced over long periods of time. For example, the
YEARS timespan database is not referenced for
output until the MONTHLY job is run--once a month.
If the high-level index of the prefix is a TSO userid or an
index that already exists, then the required catalog indexes
and master catalog pointers are already built and no further
action is required in this step. If not,
then the following steps must be performed.
The choice of Access Method Services (AMS) commands that are used to
define this high-level index depends on whether a master or
user catalog is to be used.
AMS commands for a Master Catalog index
definition:
No action required, but not recommended.
AMS commands for a User Catalog index definition
(assumes user catalog already exists):
DEFINE ALIAS(NAME('---name---') -
RELATE('user-catalog-name'))
The '---name---' field must be the high-level index of the
prefix. This index may be common to all of sharedprefix,
prefixes and tapeprefixes.
Special Considerations:
1. No action is required if the high-level index has been
defined already in the master catalog.
2. In a multi-CPU environment, define the
high-level index in a user catalog
and not allowed this index to default to the master catalog. Data
sets that are cataloged in the master catalog of one
system are not available from any other system's
master catalog unless the data sets are explicitly
cataloged on that system. The DEFINE ALIAS command should
be executed against every system's master catalog where
you want to have access to CA MICS data sets.
For example, if the prefix was DB2.PRE and the tapeprefix was
DB2T.PRE, the following definitions might be made:
If a Master Catalog index was to be used:
No action required, but not recommended.
If a User Catalog named SYS.USERCAT was to be used:
DEFINE ALIAS(NAME('DB2') -
REL('SYS.USERCAT'))
DEFINE ALIAS(NAME('DB2T') -
REL('SYS.USERCAT'))
Copyright © 2014 CA.
All rights reserved.
 
|
|