IBM Mainframe Forum

Hi all,

Automatic HSM dataset backup is enabled 4 Storage Groups since last 10 years. Also, we have set Expire after Days Non-usage as NOLIMIT in the management class attribute. So almost millions of datasets are reside in HSM tape. Now, wants to clean all the datasets(HBDELETE) with older than 1/2 years. Each datasets on the HSM tapes are different versions.

Now we are doing HLIST against all the datasets and HBEDLETE issued with help of simple REXX code. It already took too much time. Still, lot have to do. Is there other way to do with less time.

I have referred HSM Guide - DFSMShsm Managing ur own data. But not helpful. Can anyone done before ? Please share !!!!!

Antony

Have you been running the EXPIREBV utility of DFSMShsm on a regular basis to remove redundant backup versions.

Also, it is very very stupid and dangerous to delete backup versions and rely only on the ML2 copy of a dataset.

You really need to investigate the retention requirements of each and every dataset before you go around deleting things, especially if they are a legal requirement.
What happens if you delete the backup versions of a legally required dataset and the ML2 copy, if it exists, become unreadable.

You need to talk to your storage management people to resolve this.

Thanks a lot. It is already agreed by application support people. So we started. Planning to keep last 2 years datasets on HSM tapes.

Antony