Previous Topic: 3.1 Structural DescriptionNext Topic: 3.3 Resource Element Files


3.2 Usage Guidelines

 
Resource element files contain data elements you select from
CA MICS files or user-defined data elements computed from
CA MICS data elements.  For example, you might construct a
hardware utilization profile of the system that contains CPU,
LCU, expanded storage, and device activity data.  Since an
important issue in workload forecasting is the tracking and
verification of previous forecasts, you could obtain a
similar set of observations some weeks or months later to use
in evaluating the previous forecast.

Meta files contain data elements you select from already
defined Capacity Planner files or user-defined data elements
computed from other data elements. For example, you might
create a meta file that combines information from two
different standard application files, containing information
from the MVS/ESA Standard Application in conjunction with
data from the VAX/VMS CPU Planning Standard Application.

Business element files contain user-specified and supplied
data elements from sources other than CA MICS files. For
example, you might construct a workload profile of the system
containing the number of general ledger accounts, the number
of accounts updated, the number of reports generated, etc.
You can directly specify forecasts of future workloads or
they can be generated by the CA MICS Capacity Planner's
forecasting tools.

Since different versions of each of these files can be
created over a period of time, the product creates a file
definition from the parameters you supply on the control
screens.  This definition is stored as a file definition
inquiry in a standard MICF catalog.  You can create
additional copies or variations of the file by repeating,
modifying, and executing the cataloged inquiry.  This
eliminates the need to retype all of the control parameters
each time you create another file.

Figure 3-2 presents a schematic representation of the
creation of a capacity planning database resource element
file.  You can specify any number of standard CA MICS files
to use as input in creating and updating the capacity
planning database through parameters you specify on the
online screens.  You can identify the data elements that you
wish to extract from each file, along with data selection
criteria and the level of summarization of the data.

Each capacity planning database file can have, as part of the
summarization sequence, up to nine, user-defined data
elements, CAPAPUs, which are the application workload units.
You can use these data elements to provide installation-
specific summarization of the data.  The selected data
elements from each of the specified CA MICS files are then
merged together.  You can also have user-defined data
elements which are computed from the CA MICS data elements.
The capacity planning database files serve as input to the
various forecasting and analysis programs which are a part of
this product.
 
                  Primary            Secondary
Input File        Summarization      Summarization
TimeSpan          Level              Level
 
DETAIL            DAY                ENDTS/DAYNAME/HOUR/ZONE
DETAIL            WEEK               HOUR/ZONE
DETAIL            MONTH              HOUR/ZONE
DAYS              DAY                DAYNAME/HOUR/ZONE
DAYS              WEEK               HOUR/ZONE
DAYS              MONTH              HOUR/ZONE
WEEKS             WEEK               HOUR/ZONE
MONTH             MONTH              HOUR (if available)/ZONE
 
 
Note:  When the secondary summarization level is END TIME
STAMP, the files in which the data elements reside cannot be
merged.  For example, if you select DETAIL data with a
summarization of ENDTS, you can only select one CA MICS file
as input.  Also note, ENDTS ia a required data element.


+--------------+ | | | CONTROL | | PARAMETERS | | FROM | | SCREENS | +--------------+ | | G E N E R A T E D | | S A S C O D E +-----------------------------------------------------+ +--------------------------------------------------+ | | | *********** *********** *********** *********** | V * CA MICS * * CA MICS * * CA MICS * * CA MICS * V *Database * *Database * *Database * ...for each file you define *Database * *********** *********** *********** *********** | | | | +-----------+ +-----------+ +-----------+ +-----------+ User ==>| Selection | | Selection | | Selection | | Selection | Specifications +-----------+ +-----------+ +-----------+ +-----------+ | | | | +-----------+ +-----------+ +-----------+ +-----------+ | Sort | | Sort | | Sort | | Sort | +-----------+ +-----------+ +-----------+ +-----------+ | | | | +-----------+ +-----------+ +-----------+ +-----------+ User ==>| Summary | | Summary | | Summary | | Summary | Specifications +-----------+ +-----------+ +-----------+ +-----------+ | | | | +-----------+ +-----------+ +-----------+ +-----------+ | Merge Sort| | Merge Sort| | Merge Sort| | Merge Sort| +-----------+ +-----------+ +-----------+ +-----------+ | | | | +----------------------------------------+ +-----------------------------------+ | | V V +-------------+ User ==>| Merge | Specifications +-------------+ | V *********** * * * CAPACITY * * PLANNING * * DATABASE * ***********


 Figure 3-2.  Capacity Planning Database Creation Flow Chart

The CA MICS Capacity Planner's online facility introduces the
concepts of application workload unit and forecast age.
These are important concepts to understand if you wish to
take full advantage of the product's capabilities.

APPLICATION WORKLOAD UNIT (CAPAPU)
 
In many capacity planning applications you may find it useful
to group observations from the input CA MICS files into
logically similar application or workload groups.  For
instance, in tracking DASD use at your installation you might
find it more meaningful to track the use of groups of volumes
(for example, SYSDA volumes or TSO online volumes) rather
than to track the use of each separate volume.  You can
perform this task through the application workload unit
feature.

You can implement the application workload unit feature by
adding up to nine, optional, 64-byte character data elements
(CAPAPUs) to the capacity planning database historical and
forecast files.  This feature is implemented in your
individual capacity planning database files only if you
specify it.  If you specify CAPAPUs, you are responsible for
specifying the SAS statements to perform the application
workload groupings and for defining the corresponding values
of the CAPAPU data elements.

FORECAST AGE
 
For any numeric data element in your capacity planning
database files, it is possible to generate and track multiple
forecasts that are each based on a different historical range
of data.  The method that the capacity planning forecast
files uses for identifying the separate forecasts is the
inclusion of two data elements: FORYEAR, and FORMNDY,
FORMONTH, or FORWEEK (depending on the primary summarization
level of the historical data used to generate the forecast).
These data elements identify the most recent historical
observation used to generate the forecast.  Each observation
forecasted during a single execution of a workload
forecasting routine will contain the same values for these
two data elements.

For example, if, when using the DASD Planning Standard
Application, you generate a six-month forecast for the
BDPTOTL data element (total space used) that is based on
historical data ranging from January 1997 through June 1998,
the value of FORYEAR would be 98 and the value of FORMONTH
would be 6 for each of the six forecasted observations.

If you then produce a second set of six-month forecasts based
on historical data ranging from January 1998 through January
1999, the value of FORYEAR would be 99 and the value of
FORMONTH would be 1 for each of these six forecasted
observations.  Both sets of forecasted observations are
available in the CPFxxx01 cycle for tracking and reporting
through the Presentation Graphics routines of the CA MICS
Capacity Planner.

Whenever you save new forecasts, the CPFxxx01 cycle of your
forecast files is scanned to determine if any forecasts
existing in the file are older than the forecast retention
value that you specified when creating the capacity planning
database.  If so, they are dropped from the file.  If you
specified that forecast aging was to occur, then these
forecasted observations still exist in your older cycles of
the forecast files.  Aging of the 01 cycle into the 02 cycle
occurs prior to this scanning and only the 01 cycle of your
forecast files is scanned for forecast deletion.