OPQRYF .vs. FMTDTA

[Scott Cornell <CORNELLS@xxxxxxxxxxxxxxx>]

Rajee,

In a message dated 97-05-27 09:11:31 EDT, you write:

In the application we are using there are many CLPs with FMTDTA. now
we are planning to change all these instants by OPNQRYF statement.
We donot  know much about FMTDTA (..Especially the sort
specifications used..). Can anybody let us know how to go about with
it.
                           --------

My .02$ -

A)  I vaguely recall an article in News 3X/400 (or maybe MidRange
Computing) a couple years back comparing FMTDTA .vs. OPNQRYF
performance.  Unfortunately, I can't find the exact issue - I do think the
results were surprisingly for OPNQRYF, e.g. it performs better than
people think, but one should still be wary of using it over large files.

B) IMHOP, I'd gladly take a "slight" performance hit to replace the cryptic
FMTDTA specs with the more readable OPNQRYF QRYSLT parameter,
as well as putting the sort/select logic in the same source member as
the rest of the CLP processing.  I leave it to you to determine what
constitutes "slight" :)

C) As a practical example (for what it's worth), one extract program I
wrote does use OPNQRYF (actually CPYFRMQRYF to a QTEMP file) to
whittle down a couple of pretty substantial files (700+K records) to
something a little more manageable for further massaging.  While I'm
sure it's not a speed demon, it works acceptably well - batch
processing times even during high system usage don't exceed 5-10
minutes.
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* This is the Midrange System Mailing List!  To submit a new message,   *
* send your mail to "MIDRANGE-L@midrange.com".  To unsubscribe from     *
* this list send email to MAJORDOMO@midrange.com and specify            *
* 'unsubscribe MIDRANGE-L' in the body of your message.  Questions      *
* should be directed to the list owner / operator: david@midrange.com   *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *