The calculation of the number of records for the task (in this case I will discuss the CPU Utilization and for Linux ) for "Total Measurements" is coming from Tivoli Data Warehouse.
The blog explains how this came about and what are the other possible measurements that can be obtained in the Performance Analyzer
A snapshot of the TEP Client showing the Performance Analyzer tasks is shown below.
To double check this, login to the Database ( in this case DB2 ) and run this query for CPU Utilization for Summarized data for Hourly.
connect to WAREHOUS
select count(*) from itmuser."Linux_CPU_Averages_H"
But before this - one has to check to what granularity has the user configured the task measurements in the PA configuration window.
For this - open the Performance Analyzer configuration and view the granularity. (here it is set to Hourly - so the query will be
Linux_CPU_Averages_H
Had the granularity been set to Daily - then the query would have a suffix of "D", Likewise a granularity set to "Weekly, then the query would have a suffix of "W" and so on.
No comments:
Post a Comment