Almost by definition, DFHSM doesn't do high priority work. Even recalls. If a dataset is needed quickly, why did the migration rules allow it to migrate.
I run DFHSM in STCLOW > -----Original Message----- > From: IBM Mainframe Discussion List <[email protected]> On > Behalf Of Arye Shemer > Sent: Monday, April 27, 2020 4:27 AM > To: [email protected] > Subject: DFSMShsm high CPU consumption > > Hello forumers, > IHAC complaining about high CPU consumption by DFSMShsm. > Customer has no HSM experts and saying that while running HSM backups, > Recycle, and several other HSM activities the CPU consumption rising to > 100%. > This high CPU consumption interfere with other higher importance online > jobs and cause timeouts. > Customer is willing to let the HSM activities (backups and maintenance) take > longer in order to reduce the CPU consumption. > Which HSM parameters we should investigate in order to achieve this goal. > I assume that WLM is also need to be looked at to reduce the priority of the > HSM tasks. > Your directions and sharing your experience would be in great help, Thank > you, Arye Shemer. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
