Re: UP vs UF

["James W. Kilgore" <qappdsn@xxxxxxxxxxxxx>]

Barbara,

I learned the same, update files are record at a time.  There would be no
blocking benefit of cycle programming.  In the cycle, each and every record 
would
have to be locked, updated or not.

A technique that I have found to work quickly for a batch process is to use the
physical (unkeyed) file as IP and the index as UF and chain to those records 
that
-need- an update.

This works well when creating an OPNQRYF front end would be over complex for the
problem at hand.  IMHO a group if IF's in RPG sure bets trying to conditionally
build a QRYSLT string in CL.  You can still throw in a simple OPNQRYF select
process to ignore the bulk of a file.

bmorris@ca.ibm.com wrote:

> boothm@earth.goddard.edu wrote:
> > Does Barbara have any info to share on UP to a sequential file vs. a
> > roll-your-own-cycle job?
>
> Booth,
> There's very little (if any) difference in performance between UP
> and UF.  There's no magic in the RPG cycle that makes it able to
> read and update records any faster than a loop with READ and UPDATE
> or EXCEPT opcodes.  Same with IP/IF (as long as the IF file is
> blocked).
>
> I just ran a little test and coding my own loop to read and update
> was a tiny bit faster (40 seconds for the cycle and 39 seconds for
> my code).  This is insignificant as far as saying "don't use the
> cycle"; rather it indicates that choosing between UP and UF should
> be based on considerations other than performance.
>
> Barbara Morris
>

+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---