15. WORKLOAD CHARACTERIZATION › 15.6 Case Studies › 15.6.1 Batch Job Class Study › 15.6.1.2 Control Parameters
15.6.1.2 Control Parameters
Figure 15-18 illustrates the Workload Characterization
screen.
/------------ Workload Characterization Using Neugents Technology -----------\
|Command ===> |
| |
|Composing CA MICS Inquiry: CLASS - Batch Job Class Case Study |
| |
|CA MICS file ===> JOB - |
|Time-span: DETAIL Cycle(s) ===> 01 - 10|
|Data base ===> A _ _ _ _ |
|Allow Run-time Override of Data Base ===> Y (Y/N) and/or Cycle(s) ===> Y (Y/N)|
| |
|Analysis based on ===> JOBTCBTM TAPEUSE_ ________ ________ ________ |
| ________ ________ ________ ________ ________ |
|Report for each cluster ===> JOBEDASD JOBETAPE JOBTOHRS JOBNLR ________ |
| ________ ________ ________ ________ ________ |
| |
|Data Selection: Allow run-time override|
| Date ===> _______ - _______ _______ - _______ ===> N (Y/N)|
| Hour ===> __ - __ __ - __ __ - __ ===> N (Y/N)|
| Zone ===> 1 2 3 4 _ _ _ _ _ ===> N (Y/N)|
| SYSID ===> SYSA ____ ____ ____ ____ ____ ===> N (Y/N)|
| |
| |
|Specify extended options ===> Y (Y/N) |
\------------------------------------------------------------------------------/
Figure 15-18. Workload Characterization Screen Control Parameters
The Workload Characterization screen contains the following
fields:
CA MICS The CA MICS file to be used for workload
file: characterization. (For this example, we
specified the CA MICS BATJOB File for a job
class study.)
Cycles: The number of cycles to be specified. (For
example, we specified ten cycles. In an actual
study, analyze as much data as possible before
you initiate a new job class structure.)
Database: The single-character database identifier for
this file at your site.
Analysis The CA MICS data element feature names (CPU
based on: Time and Tape Drive Usage) to be used in the
clustering procedure. Because TAPEUSE is a
user supplied element that is used to summarize
tape device usage, it is defined on the
Non-CA MICS data derivation exit panel and uses
the DRVMAC macro to populate the data values.
(As discussed in Section 15.2.4, you should
specify as few features as possible.)
DASD EXCPs, Tape EXCPs, Tape Drive Occupancy
Time, and print lines are specified for
reporting purposes.
/--------------------------- Extended Options Menu --------------------------\
|Command ===> |
| |
|Inquiry Step: Workload Characterization |
| |
| |
|Specify global data selection exit ===> N (Y/N/R) |
|Specify non-CA MICS data element derivation exit===> Y (Y/N/R) |
| |
|Specify alternate input file ===> N (Y/N/R) |
|Save selected input observations ===> N (Y/N/R) |
|Save cluster output observations ===> Y (Y/N/R) |
| |
|Specify data filtering parameters ===> Y (Y/N/R) |
|Specify clustering execution options ===> Y (Y/N/R) |
|Define report description element names ===> N (Y/N/R) |
| |
| |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-19. Workload Characterization Extended Options Menu
You may select additional option specifications on the
Extended Options Menu. Figure 15-19 shows an example with
four of the extended options specified.
/------------------------------- SAS Statements -----------------------------\
|Command ===> Scroll ===> CSR |
| |
|Inquiry Step: Workload Characterization |
| |
|Line Cmds: I Insert D Delete R Repeat M Move C Copy |
| |
|Cmd SAS Statements for: Specifying data element derivation |
| - ------------------------------------------------------------------------ |
| _ - %MACRO DRVMAC; |
| _ IF JOBMXNTA EQ 0 THEN TAPEUSE = 0; |
| _ ELSE IF 01 LE JOBMXNTA LE 3 THEN TAPEUSE = 1; |
| _ ELSE TAPEUSE = 2; |
| _ LABEL TAPEUSE = "Tape Device Usage Indicator"; |
| _ - %MEND DRVMAC; |
|****************************** BOTTOM OF DATA *******************************|
| |
| |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-20. User-Defined Data Element Derivation Macro
The SAS statements that compute the user-defined data
elements are placed in the DRVMAC macro shown in Figure
15-20.
/--------------------- Save Cluster Output Observations ---------------------\
|Command ===> |
| |
|Inquiry Step: Workload Characterization |
| |
|SAS file name ===> ALLOBS |
| |
|SAS data set name ===> 'your.dataset.name' |
| |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-21. External Report Output File Definition
In this case study we wanted to create the SAS data set name,
ALLOBS, that contains the cluster analysis output. To do
this, we specified ALLOBS as the SAS file name on the
External Report Output File Definition screen shown in Figure
15-21.
/--- Data Filtering Options For Neugents Technology ---- ROW 1 OF 1\
|Command ===> Scroll ===> CSR |
| |
|Inquiry Step: Workload Characterization |
| |
| Please enter the names of elements that should be checked for zero |
| values. Observations with elements specified below that contain zero |
| values will be subsequently discarded from further processing. |
| |
|Line Cmds:%I+Insert %D+Delete %M+Move |
| |
|Cmd Element Data element label (long name) Warning |
| - -------- ---------------------------------------- -------- |
| _ JOBTCBTM |
| _ |
| _ |
|****************************** BOTTOM OF DATA *******************************|
| |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-22. Data Filtering Options
We excluded jobs that used no CPU time from the study, since
they did not execute due to a JCL error, or some other
abnormal condition. To exclude jobs with no CPU usage,
specify the data element name, JOBTCBTM, on the Data
Filtering Options for Neugents Technology screen shown in
Figure 15-22.
/---------- Execution Parameters for Neugents Technology --------- ROW 1 OF 5\
| |
|Inquiry Step: Workload Characterization Using Neugents Technology |
| |
| Clustering Input Data ===> SAMPLE (SAMPLE/POPULATION) |
| Clustering Method ===> RADIUS (RADIUS/NUMBER OF CLUSTERS) |
| Maximum Cluster Radius ===> 3.0 (1.0-3.5) If RADIUS Specified |
| Target Number of Clusters ===> (03-64) If CLUSTER Specified |
| |
| Training Sample Size ===> 2000 (nnnnnn/ALL) |
| Sample Trim Limit ===> 02.5 (0.00-99.9) Default: 2.5 Pct. |
| |
| Include Outliers in Clusters ===> YES (YES/NO) Trim 02.5 Pct. |
| Include Sparse Clusters ===> YES (YES/NO) LT 0.5 Pct. of Obs. |
| |
| Report Cluster Population ===> YES (YES/NO) |
| Report Using Account Fields ===> YES (YES/NO) |
| Report Outliers Separately ===> YES (YES/NO) |
| Report Sparse Clusters ===> NO (YES/NO) |
| Create Cluster Index Graph ===> NO (YES/NO) |
| Create Population Graph ===> NO (YES/NO) |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-23. Execution Parameters Screen
Using the default execution parameter values for the study
will normally provide good results, especially for a first
time execution. For this case study, we will generate the
Cluster Population Summary and Data Value Exception Detail
reports, and include both outliers and sparse clusters in
both the analysis and in all reports. This should provide a
good overall picture of the data under study and help
determine how to apply the clusters in the job class
structure. All of these options are presented in Figure
15-23 above.
/------------------------ Report Description Elements ----------- ROW 1 OF 2\
|Command ===> Scroll ===> CSR |
| |
|Inquiry Step: Workload Characterization |
| |
| Note: The application will automatically insert all key sequence |
| elements relevant to the CA MICS file being analyzed into |
| the clusters for use in optional reporting. There is no need |
| to include any key sequence elements. |
| |
|Line cmds: I Insert D Delete M Move |
| |
|Cmd Element Data element label (long name) Warning |
| - -------- ---------------------------------------- -------- |
| _ JOBCLASS |
| _ |
|****************************** BOTTOM OF DATA *******************************|
| |
| |
\------------------------------------------------------------------------------/
Figure 15-24. Report Description Elements Screen
Since we wished to create an optional external report data
set during the study, we also specified two descriptive
variables. We selected the variable JOBCLASS (job class), so
that the results of the cluster analysis could be compared
with the current job class structure and the assignment of
particular jobs could be determined. These data elements are
specified on the Report Description Elements screen shown in
Figure 15-24.
/----------------------------- Inquiry Processing ----------------------------\
|Command ===> |
| |
| |
| |
| Active MICF Inquiry: CLASS - Batch Job Class Case Study |
| Author: Charles Frost |
| |
| |
| Specify Inquiry processing ===> EXECUTE (EXECUTE/MODIFY/SAVE/CANCEL) |
| EXECUTE - Execute the inquiry. |
| MODIFY - Redisplay the inquiry for modification. |
| SAVE - Exit from inquiry composition and save the inquiry. |
| CANCEL - Exit from inquiry composition without saving the inquiry. |
| |
| |
|Instructions: |
| |
| Specify desired option and press ENTER key. |
| |
| |
| |
| |
\------------------------------------------------------------------------------/
Figure 15-24A. Inquiry Processing
After a workload characterization analysis inquiry is
created, the Inquiry Processing screen shown here is
displayed. From this screen, you can execute the newly
created inquiry, modify it, save it without executing it, or
cancel the parameter specifications you just entered.
Copyright © 2014 CA.
All rights reserved.
|
|