The figures below show the Relative Importance Analysis Reports for this case study. As this report shows, there is no significant difference between the 'average' and 'maximum' data element values and in both groups, BATCH (CPUxxB) is the least relevant. The other workloads, are about even in terms of affect on the CPUTOBTM value. The BATCH workload value was retained, and the 'maximum' data values were used in the second execution.
Resource Component Analysis Report Report For: 01/28/04 Feature Description Minimum Maximum Average Std. Dev. CV Rel. Imp. ________ ________________________________________ ____________ ____________ ____________ ____________ ______ _________ CPUTOBTM Total Time Processor Busy 0:08:15.73 0:27:00.70 0:19:23.79 0:03:23.89 17.52 CPUAVB Avg Batch Users 3.16 10.23 5.16 1.83 35.50 0.1465 CPUAVE Avg OMVS Users 12.00 12.10 12.03 0.02 0.15 0.2229 CPUAVS Avg Started Users 196.47 221.82 206.65 7.42 3.59 0.1482 CPUAVT Avg TSO Users 2.04 16.72 9.96 4.15 41.71 0.1537 CPUMXB Max Batch Users 5.00 17.00 8.76 2.85 32.50 0.0846 CPUMXE Max OMVS Users 13.00 13.00 13.00 0.00 0.00 0.0000*** CPUMXS Max Started Users 199.00 224.00 208.87 7.73 3.70 0.1126 CPUMXT Max TSO Users 3.00 19.00 11.73 4.55 38.78 0.1315 Total Relative Importance Weights: 1.0000 *** - This Field(s) is 100% Correlated to the CPUTOBTM Value and was Dropped from Analysis.
Figure 13-1.09. Relative Importance Analysis Report - first execution For the second execution, only the 'average' (CPUAVx) data values was used. Not surprisingly, the approximate ratios of the Relative Importance Index value didn't change much between executions. In a real-life study, the actual processor times for the workloads would need to be considered as well as the simple number of users, and a detailed workload characterization study would be required before using the metrics for any forecasting operation.
Resource Component Analysis Report Report For: 01/27/04 Feature Description Minimum Maximum Average Std. Dev. CV Rel. Imp. ________ ________________________________________ ____________ ____________ ____________ ____________ ______ _________ CPUTOBTM Total Time Processor Busy 0:08:15.73 0:27:00.70 0:19:23.79 0:03:23.89 17.52 CPUAVB Avg Batch Users 3.16 10.23 5.16 1.83 35.50 0.2265 CPUAVE Avg OMVS Users 12.00 12.10 12.03 0.02 0.15 0.2834 CPUAVS Avg Started Users 196.47 221.82 206.65 7.42 3.59 0.2333 CPUAVT Avg TSO Users 2.04 16.72 9.96 4.15 41.71 0.2568 Total Relative Importance Weights: 1.0000
Figure 13-1.10. Relative Importance Analysis Report - second execution
|
Copyright © 2014 CA.
All rights reserved.
|
|