---------------------------------<snip>-----------------------
From my little knowledge, I prefer not cataloging more datasets than
necessary in the master. I don't know if this is a wrong positioning. Is it?
Nevertheless, my main doubts are the catalogs I should use...
---------------------------------<unsnip>---------------------
Reserving the Master Catalog for the operating system, and as few other
datasets as possible, is a good idea. As far as other catalogs are
concerned, I like a dsname standard that separates PROD from
TEST/DEVELOPMENT as early as possible in the name. That is, in the
qualifier as far to the left as possible, ideally in the first
qualifier. Then I can direct PROD to one catalog, or one small set of
catalogs, and keep all other activity away from it. So a programmer that
might mess up a catalog can't affect my PROD workload. Remember that
PROD is most often the revenue-producing part of the IT environment;
protect it. Don't go overboard defining catalogs; maintenance can get to
be a real headache and disaster recovery can be adversely affected. I've
worked mostly in small shops and considered that 3-4 USERCATALOGS was
more than sufficient (PROD, DataBase, Systems Programming and all
other). YMMV.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
