13. RESOURCE COMPONENT ANALYSIS › 13.2 Analysis using CAP Database files › 13.2.6 Case Study › 13.2.6.1 Processor Resource Usage

13.2.6.1 Processor Resource Usage

This case study shows how you can query a set of workloads
and determine which workloads most significantly affect the
processor resource usage.

In preparing for a capacity planning forecast, it is
important for you to understand which workloads have the
greatest impact on processor utilization. While you might
want to simply look at workload CPU time, there may well be
relationships among workloads that are not readily apparent.
A better approach is to use Neugents technology Relative
Importance, which is a nonlinear process and is therefore
more aware of such relationships.

This case study assumes that two files exist in the capacity
planning database, one built with the CPU Planning Standard
application using the CPU Processor Activity (HARCPU) data as
input, and another built as a Resource Element file. The
Resource Element is derived from the Service Class Resource
Consumption (WLMSEC) data. Both files in this sample reside
in the CA MICS Hardware and CSP Analyzer (RMF) area and are
summarized at the "Weeks" timespan. The WLM workload usage
data has been mapped to a series of client defined workload
names and the file contains one set of workload data for each
observation.

DEVELOPMENT OF CONCEPTUAL STUDY

This study is being conducted as a prelude to performing a
capacity study of a single system using 53 weeks of input
data. The workload measurement data will be analyzed with
respect to the processor resource usage in an effort to
identify those workloads that are most relevant to future
forecasting efforts. This study will be performed initially
using all of the workloads, and then executed again after
removing (unselecting) those workloads that appear to be less
relevant.

APPLICATION OF CONCEPTUAL STUDY

The system being examined is SYS1 (this is the CA MICS SYSID
value).  This system processes a varied workload, which can
be divided by WLM Service Class and summarized by Workload
Name. The workloads are a TSO subsystem, a Database
subsystem, an Online subsystem, a BATCH subsystem and a small
USS based workload.  There are also system-related tasks that
must execute to provide an array of common services, such as
GRS, JES2, VTAM, TCAS, and RMF.

Two files contain CPU time information concerning these
workloads and processor use as a whole.  These files are:

    APU - Processor Usage File
    WKL - Workload Usage File

In our example, the processor usage for each workload is
analyzed against the total processor usage and the effect
quantified into a weight factor.

The next sections explore the results of the report that is
generated from the Resource Component Analysis we developed
for workload usage study:

     1 - Control Parameters
     2 - Resource Component Analysis Report