Previous Topic: History Records Compared to Account Records

Next Topic: Considerations When Using a Summarized History File

Summary File Examples (Levels 1-5)

Summarized history files are created after the sort and prior to the creation of the output data elements table.

Again, the records are in the basic account record format. A level-5 file contains the most records and level-1 the least.

  1. File level (field position 2) is a 1-5 and be the same as the history-level flag on the SORT statement.
  2. Processing ID (field position 28) is an H.
  3. Sysout table is absent. Total standard lines, total special lines, total lines printed and pages printed are retained in the record.
  4. The I/O device table is not created. Total I/O counts by device type and total counts of the individual device types are present in the record.
  5. All fields which are additive or capable of being mathematically accumulated are present and reflect the totals from the records in that sort break.
  6. Since character fields are not cumulative, they are usually set to blanks, unless the field was one of the sort keys at the time the file was created.
  7. EXTDATA is absent.

Example:

1st sort level is job start date (month and year only).

2nd sort level is a 4 position accounting field that reflects departments.

3rd sort level is job name.

A level-3 history file contains one record for every unique job name run for an individual department for specific month and year.

If job A has been run 4 times for the accounting department in March of 1988, then one record is written to the history file containing all the accumulated statistics for four executions of job A. It also contains the accounting information and month/year values.

A level-2 history file, based on the above information, has one written for each department for a specific month and year. The accumulated statistics are for all jobs run for that department, in that month and year. Therefore, in contrast to the level-3 history file, a level of detail is missing because the file does not have statistics for each unique job name within the department.