RE: SQL vs native access (was: Record name the same as the filename)

Is there ever a condition under which SQL will be faster than Input/Primary?

 
 
 
---------------------------------------------------------
Booth Martin http://www.MartinVT.com
Booth@xxxxxxxxxxxx
---------------------------------------------------------
 
-------Original Message-------
 
From: RPG programming on the AS400 / iSeries
Date: Thursday, August 21, 2003 9:15:20 PM
To: 'RPG programming on the AS400 / iSeries'
Subject: RE: SQL vs native access (was: Record name the same as the
filename)
 
> From: MEovino@xxxxxxxxxxxxxxxxx
>
> Joe,
>
> Just curious here (in no way impugning your tests), what versions of
the
> OS
> have you run your tests on? I wonder how that snappy new query engine
in
> V5R2 performs.
 
Haven't run them yet on V5R2, Mike. My last tests were on V4R5. But
I'd be flabbergasted to learn though that a CHAIN from a native record
would be slower than a SELECT/FETCH (there's just more overhead on the
latter). The problem though is that in order to be fair I really need
to use the fastest syntax available and that would require working with
a good SQL expert. The first step would be to at least set up a set of
criteria that we could all agree on to compare performance.
 
Remember, though, that queries, especially bulk queries, were never the
issue. Single record fetches were, and more importantly single record
inserts and updates - the sorts of things we do on a daily basis in
typical business applications. I'm willing to wager that those are
still faster with native I/O.
 
However, my gut feeling is that at this point SQL may have caught up
enough that in many cases the difference would be that "stepladder to
get closer to the moon". I just don't have the time to make that
comparison.
 
Joe