While working at multiple sites I have seen customers activating the DLF
address space often. DLF supports a facility known as Hiperbatch, which was
introduced in the early 1990s. Hiperbatch is a way of making data available
in expanded storage from one job step of a batch job to another job step,
thereby reducing I/O. (Hope you all remember expanded storage folks.) The
manual for Hiperbatch is no longer distributed with those for z/OS but is
still available. It is dated 1994 and is still at Version 1 I believe. DLF
is controlled from the COFDLFxx member of PARMLIB. I understand there are 3
ways to make a dataset eligible for Hiperbatch assistance, 

1.      Through a RACF profile in the DLFCLASS class
2.      Through the DLF exit (specified in the COFDLFxx member)
3.      Through the OPC/ESA (long since renamed) special resources dialog,
also using a DLF exit.

 

I think that if DLF is active and the DLFCLASS class is active then each
data set OPEN will cause an extra RACF check to the DLFCLASS class to see if
the data set is eligible. I have seen these checks in zSecure Access
Monitor.

 

So I have two questions. Number 1 is for most people.

Is anyone still using Hiperbatch?

 

Question 2 is for IBM.

In many cases we can simply not start DLF and inactivate the DLFCLASS class
and thereby save processing cycles. What is the downside to this?

 



Lennie Dymoke-Bradshaw

 <https://rsclweb.com/> https://rsclweb.com 


'Dance like no one is watching. Encrypt like everyone is.'

 

 

 

 

 

 


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to