Job end information - syncsort_capacity_management - 12 - 12.40

This comes from SMF type 30 subtype 5 records and like their name suggests, are cut by z/OS when a whole job ends. When Control Center reads a job end record from Acquire it will apportion usage of CPU, I/O etc back across the RMF intervals where the job was running, in proportion to its elapsed time and the RMF interval length. Here is an example:

JOB1 starts at 21:33:24.17 and finishes at 22:29:43.54, an elapsed time of 00:56:19.37, or 3379.37 seconds. The job used 33.14 seconds of CPU time.

The RMF intervals are each 30 minutes (1800 seconds) long starting at 21:30 22:00 and 22:30

Control Center will take select the RMF intervals where the job was active and apportion the CPU time across them like this:

21:30 - 22:00 Job active for 00:26:35.83, or 1595.83 seconds.

CPU = 1595.83/3379.37 * 33.14 = 15.65 CPU seconds

22:00 - 22:30 Job active for 00:29:43.54 or 1783.54 seconds

CPU = 1783.54/3379.37 * 33.14 = 17.49 CPU seconds

When graphed, the CPU usage looks like this:

With a job that has run for a very long time, this apportioning can lead to an artificial looking usage graph as in the example shown below:

This is unlikely to be correct in any given interval, but the resources used by the job are correct when summed over time.