11. WORKLOAD CHARACTERIZATION › 11.6 Case Studies

11.6 Case Studies

The Case Studies Section presents case studies to illustrate
how you can apply the clustering methodology to solve
capacity planning and performance management problems that
many data processing installations commonly encounter.

The batch job classes case study shows how you can use
clustering to investigate patterns in the batch jobs being
processed.  You can use information to develop workload
classes which can then form a reasonable basis for service
objectives.  The statistical uniformity of the classes
enables you to assess the degree to which the service
objective can be met.  You can use the same information to
define a job class structure for establishing an initiator
scheme that is consistent with the service objectives.  A
full set of reports and an explanation for each report is
provided with this study.

The TSO response study shows how you can apply clustering to
a TSO workload to establish a characterization, which you can
used to develop a benchmark system.  The basic objectives of
the study are presented through Descriptive Statistics,
Cluster Centroids, and Cluster Assignment Reports.

The IMS transaction study looks at an IMS workload to
determine a transaction class strategy.  The study includes a
description of the input data and the centroid report.

The case studies are shown in the following sections:

 1 - Batch Job Class
 2 - TSO Response
 3 - IMS Transaction