From: | Mikko Tiihonen <Mikko(dot)Tiihonen(at)nitorcreations(dot)com> |
---|---|
To: | Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com> |
Cc: | List <pgsql-jdbc(at)postgresql(dot)org> |
Subject: | Re: Change of format of returned flat value after prepareThreshold |
Date: | 2015-10-19 12:30:59 |
Message-ID: | DB4PR07MB049539C659EEDB4D8587F520F23A0@DB4PR07MB0495.eurprd07.prod.outlook.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-jdbc |
> > but this jdbc driver feature is
> >supposed to be more of a debug feature, because it slows down queries due to needing an extra
> >round-trip for each query.
>
> Mikko, binary transfer should not be considered a "debug" feature.
I know. I wrote the original jdbc driver binary transfer code because I wanted better performance.
But you clipped half of my sentence it stars with:
"An alternative is to request binary transfers from the first query" - meaning that the debug feature
is the performance killer that requests binary transfers for also the first query by doing a backend
round-trip to describe what kind of in/out parameters the operation contains - before executing the
actual statement.
It is currently used by jdbc driver unit test framework so that every operation need not be tried n
times in unit tests to verify that the binary transfers work for all operations.
-Mikko
From | Date | Subject | |
---|---|---|---|
Next Message | Dave Cramer | 2015-10-19 12:47:51 | Re: Change of format of returned flat value after prepareThreshold |
Previous Message | Vladimir Sitnikov | 2015-10-19 10:50:52 | Re: Change of format of returned flat value after prepareThreshold |