Appendix A. Special CA MICS Processing › A.1 CA MICS SPECIAL Database Overview

A.1 CA MICS SPECIAL Database Overview

Special CA MICS processing facilities allow the creation of a
CA MICS database called a "SPECIAL" database.  A special
database is typically used for ad hoc studies and backload
processing.  Figure A-1 shows the flow of information and
control for SPECIAL database processing.



      +-----------+
      | File      |
      | Tailoring |    +--------------+
      | Parameters|--->| SPECIAL      |
      +-----------+    | Database     |
                       | File         |
                       | Tailoring    |
                       | Generation   |
                       +--------------+
                              |
                              V
            +-----------------+
            |
            |
            V
      +------------+
      | Database   |
      | File       |-+
      | Definition | |
      | Data       | |
      +------------+ |
                     |
      +------------+ |
      | Execution  | |  +--------------+         +----------+
      | Parameters |-+->|              |         | SPECIAL  |
      +------------+    |  SPECIAL     |-------->| Database |
                        |  Database    |         |          |
      +------------+    |  Processing  |         +----------+
      | Input      |--->|              |
      | Data       |    +--------------+
      +------------+



Figure A-1. SPECIAL Database Processing

The CA MICS SPECIAL database looks much like any other
database unit installed within the CA MICS Database Complex.
There are several differences that cause the special database
to be desirable.

    - The files of a special database are not bound by
      global database timespan definitions.  The files may
      be tailored in a special way that makes it possible to
      add or drop files to/from any timespan without
      affecting the production CA MICS database units.  This
      is accomplished by user specifications in a parameter
      library member called SELFILES.  A job called SELFILES
      inputs this parameter set, and creates executable code
      to tailor files in the next database run.

    - The components processed by a SPECIAL database may be
      varied from run to run.  Also, the data selection or
      forcing of data into the database may be varied
      from run to run.  This is accomplished by user
      specifications in a parameter library member called
      EXECDEF.

RESTRICTIONS:

The SPECIAL database also has some restrictions worthy of
note.  The fact that SPECIAL databases are individually
tailored may cause some files to appear in timespans that
the component does not normally support.  Also, no archiving
is done.  Further, the SPECIAL database does not issue the
CA MICS 300 abend (the SYSPARM=NODATA option is automatically
provided).

The CA MICS SCHEDULE process has no practical use in a
SPECIAL database.  Backloading and ad hoc studies of old
data do not use the schedule process to the advantage of the
database unit.  SCHEDULE tends to package runs that execute
WEEKLY and MONTHLY processes at inappropriate times.

The CA MICS Accounting and Chargeback Product may not be
installed in a SPECIAL database.  A SPECIAL database is
designed to analyze 'data'.  To test out an accounting
method or actual implementation requires a TEST unit.

Finally, periodic changing of execution components may create
a confusing mix of file cycles.  The detail timespan, cycle
01, for file 'x' may represent data from yesterday, while the
same cycle of the same timespan for file 'y' may contain
data from several weeks ago.