A sample Cluster Population Summary report is shown below in Figure 15-3.
Workload Characterization Analysis 1 Cluster Population Summary For: Tuesday, June 24, 2003 _________________________________________________ Cluster: 1 ____________________________________________________ Radius: 0.00 Clustering Index: 0.00 Maximum Cluster Radius: 3.0 STDs Normal Obs.: 0 0.00% Observations << 97.5 Pct. Outlying Obs.: 14 0.70% Observations >> 97.5 Pct. Total Obs.: 14 0.70% Cluster Feature Resources Feature Description Minimum Average Standard Maximum Total % of Value Value Deviation Value Value Total JOBTCBTM Job TCB CPU Time 25 656 687 2,400 9,186 34.96 JOBEDASD DASD EXCPS 5,103 236,581 298,042 1,154,523 3,312,131 27.64 Cluster Report Resources Feature Description Total % of Value Total JOBNLR Total Logical Writer Records 0 0.00 _________________________________________________ Cluster: 10 ____________________________________________________ Radius: 1.61 Clustering Index: 0.34 Maximum Cluster Radius: 3.0 STDs Normal Obs.: 1,541 77.05% Observations << 97.5 Pct. Outlying Obs.: 0 0.00% Observations >> 97.5 Pct. Total Obs.: 1,541 77.05% Cluster Feature Resources Feature Description Minimum Average Standard Maximum Total % of Value Value Deviation Value Value Total JOBTCBTM Job TCB CPU Time 0 1 2 14 1,873 7.13 JOBEDASD DASD EXCPS 1 914 1,155 6,523 1,408,821 11.76 Cluster Report Resources Feature Description Total % of Value Total JOBNLR Total Logical Writer Records 19,968 99.56 Sparse Clusters Have Been Excluded _______________________________________________________________________________________________________________________
Figure 15-3. Cluster Population Summary report
Workload Characterization Analysis Cluster Population Summary For: Monday, June 23, 2003 The following clusters were determined to be 'sparse' in their population and have been excluded from processing. Cluster Population 2 2 3 6 4 2 5 2 6 4 7 3 8 2 9 5 12 9 13 2 18 2 19 2 20 3 21 2 22 8
CLUSTER Cluster numbers are assigned sequentially
NUMBER: to the patterns that are identified by the
algorithm. Note that the order in which the
clusters are identified is not an indicator of
merit.
RADIUS: The geometric distance from the cluster center
centroid to the outer boundary of the cluster,
expressed in terms of Standard Deviations. The
outer limit of this value is defined by the
analyst as the Maximum Cluster Radius value on
the Execution Options panel.
CLUSTERING Formally called the Performance Index, this
INDEX: metric was renamed in this implementation to
avoid confusion with similar terms in the z/OS
Workload Manager. It is the Root Mean Square
of the distances of all cases (observations)
within the cluster and serves as a simple
measure of clustering effectiveness. The lower
this value, the tighter the fit of the cluster
data. This is not to say that outliers are not
present; only that they are not distorting the
cluster shape by their presence.
MAXIMUM The maximum size of any cluster in the study
CLUSTER under consideration, expressed in terms of
RADIUS: standard deviations. This value is specified
by the user on the Execution Options panel and
is presented here for documenting the cluster
definitions.
NORMAL The count of observations within this cluster
OBS: where all feature values were found to be
normal in a statistical sense. In this
implementation, normal feature values are those
that are less than the value determined by the
Sample Trim Limit. For example, if the Sample
Trim Limit is 97.5%, then all feature values of
"normal" clusters will reside below the 97.5
percentile of the sample.
OUTLYING The count of observations within this cluster
OBS: where one or more feature values were found to
be "outliers" in a statistical sense. In this
implementation, outlying feature values are
those that are greater than the value
determined by the Sample Trim Limit. For
example, if the Sample Trim Limit is 97.5%,
then at least one feature value of "outlying"
clusters will reside above the 97.5 percentile
of the sample.
TOTAL OBS: The sum of NORMAL and OUTLYING observations.
CLUSTER FEATURE RESOURCES:
For each feature (clustering element) defined on the Workload
Characterization panel, a separate line is generated here and
contains the following elements:
FEATURE: The name of the data element chosen for
clustering. This is normally a CA MICS data
element, but can be a computed (user defined)
element if required.
DESCRIPTION: The SAS label of the selected data element,
from either the CA MICS GENLIB definition or
supplied by the user.
MINIMUM The minimum of all values for this feature
VALUE: within this cluster.
AVERAGE The average of all values for this feature
VALUE: within this cluster. This value approximates
the centroid value for this feature and is also
referred to as the feature mean.
STANDARD The standard deviation of all values for this
DEVIATION: feature within this cluster.
MAXIMUM The maximum of all values for this feature
VALUE: within this cluster. This value approximates
the outer boundary value for this feature.
TOTAL The sum of all values for this feature within
VALUE: this cluster.
% OF The percentage of the sum of this feature's
TOTAL: values for this cluster compared to the sum for
the entire sample population. For example, in
the report above, the JOBTCBTM represented by
cluster 1 is nearly 35% of the JOBTCBTM for the
entire sample.
CLUSTER REPORT RESOURCES:
For each feature (reporting element) defined on the Workload
Characterization panel, a separate line is generated here and
contains the following elements:
FEATURE: The name of the data element chosen for
clustering. This is normally a CA MICS data
element, but can be a computed (user defined)
element if required.
DESCRIPTION: The SAS label of the selected data element,
from either the CA MICS GENLIB definition or
supplied by the user.
TOTAL The sum of all values for this feature within
VALUE: this cluster. This value represents the total
of this reporting element for this cluster.
% OF The percentage of the sum of this feature's
TOTAL: values for this cluster compared to the sum for
the entire sample population. For example, in
the report above, the JOBNLR represented by
cluster 1 is 0% of the JOBNLR for the entire
sample.
If the INCLUDE SPARSE CLUSTERS options on panel CAPG910U has
been set to "NO" by the analyst, the page of the report will
contain a listing of all clusters excluded, and their
populations.
The following fields are presented in this section:
CLUSTER Cluster numbers are assigned sequentially
NUMBER: to the patterns that are identified by the
algorithm. Note that the order in which the
clusters are identified is not an indicator of
merit.
POPULATION: The population of each excluded cluster.
|
Copyright © 2014 CA.
All rights reserved.
|
|