Miklos Szigetvari wrote:

>We are using the default names, without any prefix, so EOX EOY for book 
>manager etc etc

Careful with that assumption. You need to look way in the future. For example, 
some of these datasets are to be cataloged in the MCAT, this could influence 
your decision.

Other problem is, naming standards across LPARS, SysPlexes, data centres, DRP, 
etc may or may not be consistent. Oh, are you going to share them in a SysPlex 
with the usual headaches of catalog management?

Versioning of your products may be a challenge, unless you have a good naming 
standards to resolve that. It is no fun trying to run 2 versions of DB2 on one 
LPAR, but it can be done.

Then there are billing issues as well as RACF, SMS, HSM, TSO, etc, issues to 
consider before you decide on a HLQ.

Then the very most important issue to consider - Are you going to have datasets 
as distributed by IBM have one set of HLQ and customized (exits, custom mods, 
Test PTF/APAR) in another HLQ?

Then you need to look at HLQ of all third party products too.

What we do, some products are starting at SYS1. For example RMF, which we start 
with SYS1.SERB*. Assembler with SYS1.ASM.*, etc.

For some, we leave the HLQ as we see/receive from CustomPac/ServerPac (SMP/E), 
example: Bookmanager at EOX.*/EOY.*, SDSF at ISF.*, etc.

We generally have unique HLQ for things which were customized, so the 'vanilla' 
datasets we get from IBM and vendors remain unchanged during usage and 
distribution accross datacentres/rolling upgrades.

I have a problem with two products with the same HLQ. RACF is my friend to 
resolve that conflict. Reread Radoslaw point nr 3.

As others said, YMMV. 

I would suggest you and your team reread that Program Directory and that book 
IBM gave to you regarding installing z/OS from your previous level to your 
newly ordered level. In one of the Appendixes there are a paragraph for each 
product with all its distributed and target dataset names with the suggested 
HLQ.

HTH!

Groete / Greetings
Elardus Engelbrecht

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

Reply via email to