The Performance Database and the Performance Warehouse

OMEGAMON for Db2 PE performance data can be loaded into Db2 tables to build a performance database or a performance warehouse.

You can load performance data into Db2 tables to create a performance database. With a performance database you can:
  • Perform additional performance evaluations. As an example you can examine the access paths of dynamic SQL statements.
  • Collect historic data.
  • Use SQL for fast and easy retrieval of data.

The performance data can come from the following data groups:

  • Accounting
  • Audit
  • Locking
  • Record traces (IFCID 22, 63, 96, 125, 172, 196, 316, 365, 376, and 401)
  • Statistics
  • System parameters
  • Batch, periodic, and display exceptions

Data can be aggregated or nonaggregated:

  • In the case of aggregated data, several records are summarized by specific OMEGAMON for Db2 PE identifiers. In a report, each entry represents aggregated data. Use the SAVE subcommand to generate a VSAM data set that contains the aggregated data. When the data has been saved, use the Save-File utility to generate a Db2-loadable data set.
  • In the case of nonaggregated data, each record is listed in order of occurrence. In a trace, each entry represents nonaggregated data. Use the FILE subcommand to generate a data set that contains the nonaggregated data.

For batch, periodic, and display exceptions, a Db2-loadable data set is automatically generated for each OMEGAMON for Db2 PE execution if EXFILDD1 is defined in your JCL. The generated data set is sequential variable-blocked and can be directly loaded into Db2.

The following figure shows how performance data is formatted and loaded for each data group.

Figure 1. How performance data is formatted and loaded
This figure is a graphical representation of how performance data is formatted and loaded.

For details about loading performance data into Db2 tables, consult the RKO2SAMP library.