11. WORKLOAD CHARACTERIZATION › 11.3 Standard Output

11.3 Standard Output

Workload Characterization produces reports that describe the
original data, show the clusters which are identified from
this data, and analyze the degree of fit with which these
clusters characterize the data.

The four primary reports are the following:

o  Descriptive Statistics Report.  This report details the
   characteristics of the variables that are specified as
   criteria for the clustering analysis.  Scaling is an
   ancillary issue introduced by numerical clustering
   algorithms.  Scaling considerations will be discussed
   under Analytic Technique Tutorial in Section 11.4 of this
   document.

o  Cluster Centroids Report.  This report details the
   characteristics of the patterns that were identified in
   the workload.  Each of these patterns statistically
   represents a subset of the installation's workload.

o  Cluster Assignment Report.  This report provides the
   resource consumption characteristics of each of the
   clusters.  For example, you can use this report to
   identify resource consumption patterns that may be
   resulting in performance or capacity problems.  Like the
   Cluster Centroids Report, this report is based on a
   statistical sample of your installation's workload.

o  Population Cluster Assignment Report.  This report
   provides the same information as the Cluster Assignment
   Report, except that the entire input data set, rather than
   the statistical subset, is included.  This report allows
   you to judge how well the characterization fits the entire
   workload.

The sections which follow describe the fields, data elements,
and format of each report.

 1 - Descriptive Statistics Report
 2 - Cluster Centroids Report
 3 - Random Sample Cluster Assignment Report
 4 - Population Cluster Assignment Report