Re: [survey] New "Stable" QueryId based on normalized query text

From: Julien Rouhaud <rjuju123(at)gmail(dot)com>
To: legrand legrand <legrand_legrand(at)hotmail(dot)com>
Cc: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [survey] New "Stable" QueryId based on normalized query text
Date: 2019-03-20 21:44:30
Message-ID: CAOBaU_YMOF_9aGwm+ryE=3KxLWfOFH3Zddr-dsMt2xNwRA1XtQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, Mar 20, 2019 at 10:18 PM legrand legrand
<legrand_legrand(at)hotmail(dot)com> wrote:
>
> On my personal point of view, I need to get the same Queryid between (OLAP)
> environments
> to be able to compare Production, Pre-production, Qualif performances
> (and I don't need Fully qualified relation names). Today to do that,
> I'm using a custom extension computing the QueryId based on the normalized
> Query
> text.

IIUC, your need is to compare pgss (maybe other extensions) counters
from different primary servers, for queries generated by the same
application(s). A naive workaround like exporting each environment
counters (COPY SELECT 'production', * FROM pg_stat_statements TO
'...'), importing all of them on a server and then comparing
everything using the query text (which should be the same if the
application is the same) instead of queryid wouldn't work? Or even
using foreign tables if exporting data is too troublesome. That's
clearly not ideal, but that's an easy workaround which doesn't add any
performance hit at runtime.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Julien Rouhaud 2019-03-20 22:04:22 Re: [survey] New "Stable" QueryId based on normalized query text
Previous Message Shaun Thomas 2019-03-20 21:40:48 Re: Re: A separate table level option to control compression