The performance data collected in a compressed binary format is expanded and analyzed in a batch run that creates the data base members used as input for the reporting phase.

•          Selected detailed performance data per interval and job ID

•          Selected job summary and job *GROUP records

•          Resource summary per collection interval (interval = 15 seconds)

•          Resource summary by job name for jobs found repeatedly

•          Resource summary for users running many jobs

•          Program call stacks by job and collection interval

•          ODP (Open data path) totals by job and collection interval

•          ODP statistics per file within job and collection interval

•          Global file data analysis per file name

•          Summary per Job ID for all jobs and tasks *)

•          Detailed resource usage by collection interval and job-id *)

  *) Information on job level is normally only kept for jobs using many resources and therefore potentially being

      candidates for optimization. All jobs can be kept by using the “Keep all detail records” option.