Re: Sever pulling a bazillion rows from IBMi via JDBC (I think)

Patrik,
Yes time may change things, and might even change the following. Decades
ago, before saving access paths became a default, we had to restore a
file. It quickly restored. However, it was almost a week before we could
use it as it took that long to rebuild the access paths.

On Wed, Sep 3, 2025 at 8:01 AM Patrik Schindler <poc@xxxxxxxxxx> wrote:

Hello Jim,

Am 03.09.2025 um 13:45 schrieb Jim Oberholtzer <
midrangel@xxxxxxxxxxxxxxxxx>:

Again, that depends.

It always does. :-) Because the destination is unclear, I can only apply
general advice.

First off, how many indexes are we discussing? Secondly are they

designed correctly so the access paths can be shared, IE create the longest
access path first then let the shorter ones use it.

[…]

The destination database is not IBM i, as far as I've understood. The
source is.

Not sure how much of your further hints are still relevant in practice,
considering the increasing ubiquity of solid state memory vs. true DASD
(Disk Arm Shaking Devices). Please note the word "practice". :-)

Creating an index is quite expensive, updating it less so usually.

Uhm. The goal set by the OP was to minimize run-time, not system resource
consumption. I know the term "expensive" as another word for "resource
intensive". What's your definition?

:wq! PoC

--
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.