Missing, incomplete, or duplicate records
To generate useful models, some minimum sets of records must be present. The most typical cases are as follows:
- Only type 70 records are present—only system-wide CPU related information will be in the generated XML model and the Visualizer file.
- Only RMF type 70, 71, 72, 73, 74, and 75 records are present—only XML model TYPE=PGSCL or REPORT can be used. The workloads that are created do not have any I/O activity.
- Only RMF type 70:75 and SMF type 30 records are present—batch applications can be created.
Duplicate RMF records for types 70, 71, 72, 74, and 75 are automatically detected and discarded. However, automatic detection is not enabled for other record types. To detect and eliminate type 30 duplicate records, you need to use the DELETEDUPREC command (described in Using-Command-filters).
Automatic detection is enabled only for the key RMF records because eliminating duplicate records can consume a lot of system resources, which can significantly increase processing time. You should request automatic detection only it if you suspect record duplication has caused erroneous results.
Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*