15. WORKLOAD CHARACTERIZATION › 15.3 Standard Output › 15.3.1 Descriptive Statistics Report

15.3.1 Descriptive Statistics Report

The Descriptive Statistics Report provides a summary of the
statistics that are calculated for the cluster features.
Figure 15-1 illustrates a sample Descriptive Statistics
Report.

                     Workload Characterization Analysis
                   Clustering Input Descriptive Statistics
                         For: Thursday, June 19, 2003

       Summary for Total Sample:
  Number of Sample Observations:     2,000
  Feature               Description                    Minimum        Maximum        Average      Std. Dev.       CV
 ________ _________________________________________ ____________   ____________   ____________   ____________   ______
 JOBTCBTM Job TCB CPU Time                            0:00:00.01     0:40:00.28     0:00:13.14     0:01:42.89     7.83
 JOBEDASD Total EXCPs                                       1.00   1,154,523.00       5,992.30      34,582.08     5.77

      Summary for Trimmed Sample:
   Number of Sample Observations:     1,919
  Feature               Description                    Minimum        Maximum        Average      Std. Dev.       CV
 ________ _________________________________________ ____________   ____________   ____________   ____________   ______
 JOBTCBTM Job TCB CPU Time                            0:00:00.01     0:01:00.81     0:00:03.74     0:00:07.71     2.06
 JOBEDASD Total EXCPs                                       1.00      38,096.00       2,505.00       4,508.48     1.80

 Figure 15-1.  Descriptive Statistics Report

 Note that the report has two sections: one for the original
 sample randomly taken from the input data; and the other,
 the sample data after trimming, using the Sample Trim Limit
 value specified by the analyst on the Execution Options
 panel.

 The Descriptive Statistics Report contains the following
 fields:

OBSERVATIONS: The number of observations selected for
              processing.

FEATURE:      The feature name.  The feature names that are
              listed in the report correspond to the features
              that are specified on the Workload
              Characterization panel.  In Neugents
              technology, this is also called a "pattern."

DESCRIPTION:  The SAS label of the selected data element,
              from either the CA MICS GENLIB definition or
              supplied by the user.

MINIMUM:      The minimum value observed for the feature in
              the sample.

MAXIMUM:      The maximum value observed for the feature in
              the sample.

AVERAGE:      The average calculated for the feature
              observations in the sample.

STD DEV:      The standard deviation calculated for the
              feature observations in the sample.

CV:           The coefficient of variation.  The CV is
              calculated by dividing the standard deviation
              by the average.