Re: help with SQL performance when using ORDER BY

Justin,
That's what I did, your recommendation, which works very well. Thank you!
I wanted the result set treated more like a distinct temp table without
actually creating one in QTEMP. I am running ad-hoc requests for a
particular transaction, etc. to validate testing, not trying to improve
production performance. The same request is near "instant" using WRKQRY but
it takes me longer to put together the pieces interactively in the UI.

On Thu, Sep 24, 2020 at 1:24 PM Justin Taylor <jtaylor.0ab@xxxxxxxxx> wrote:

So you want 100 random rows from the results, and then to sort them? That
doesn't sound like a common requirement.

Would this work?
with CTE as (
select *
from TABLE_NAME
fetch first 100 rows only
)
select *
from CTE
order by COL1


date: Thu, 24 Sep 2020 06:09:24 -0400

from: Mark Villa <iseries.4.me@xxxxxxxxx>
subject: help with SQL performance when using ORDER BY

Hi all,
I have been doing a lot of "SQL checking" recently and without breaking

out

the manuals, I am just curious what you guys do when you use ORDER BY.

With

tables greater than 10 or 20 mil rows it really tanks. Otherwise, my
responses are usually subsecond.

It makes the SQL engine (and myself) look dumb when I can't say
"select these 100 rows" then ORDER THEM, don't look at 20 million rows,
don't need anything sorted but the final set.

--
Thank You,
Mark Villa

--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at https://archive.midrange.com/midrange-l.

Please contact support@xxxxxxxxxxxxxxxxxxxx for any subscription related
questions.

Help support midrange.com by shopping at amazon.com with our affiliate
link: https://amazon.midrange.com