We are running a TSM 4.2.1.9 server under OS/390. We are experiencing
serious performance problems, and I have not been able to find any
clear explanation. We currently have two late-running client backup
sessions and a backup stgpool process running. Neither client backup
uses the storage pools involved in the storage pool backup. The storage
pool backup is sometimes freezing for tens of seconds at a time with
'query process' commands showing no current file and no increase in
file or byte counts. The buffer pool hit rate is around 97.7 percent,
which is not particularly good but does not explain tens of seconds
of inactivity. The mainframe performance analysts insist that TSM is
spending small percentages of the time waiting for processor capacity,
I/O operations, or paging operations. Where do I start looking for
the underlying cause of the problems?