Re: track_planning causing performance regression - Mailing list pgsql-hackers

From Julien Rouhaud
Subject Re: track_planning causing performance regression
Date
Msg-id CAOBaU_YJUNCPuEAciSmFBS0EJ16Qa_pDH3mLjSpUnG4FtaSbzg@mail.gmail.com
Whole thread Raw
In response to Re: track_planning causing performance regression  (Fujii Masao <masao.fujii@oss.nttdata.com>)
Responses Re: track_planning causing performance regression
List pgsql-hackers
On Tue, Jun 29, 2021 at 10:45 AM Justin Pryzby <pryzby@telsasoft.com> wrote:
>
> We borrowed that language from the previous text:
>
> | Plannable queries (that is, SELECT, INSERT, UPDATE, and DELETE) are combined into a single pg_stat_statements entry
wheneverthey have identical query structures according to an internal hash calculation
 

Yes, but here's it's "identical query structure", which seems less
ambiguous than "identical structure" as iI think one could think it
refer to internal representation as much as as the query text.  And
it's also removing any doubt with the final "internal hash
calculation".

> Really, I'm only trying to fix where it currently says "a fewer kinds".

I agree that "fewer kinds" should be improved.

>        Enabling this parameter may incur a noticeable performance penalty,
> -      especially when a fewer kinds of queries are executed on many
> -      concurrent connections.
> +      especially when queries with identical structure are executed by many
> +      concurrent connections which compete to update a small number of
> +      pg_stat_statements entries.
>
> It could say "identical structure" or "the same queryid" or "identical queryid".

I think we should try to reuse the previous formulation.  How about
"statements with identical query structure"?  Or replace query
structure with "internal representation", in both places?



pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: Pipeline mode and PQpipelineSync()
Next
From: Bharath Rupireddy
Date:
Subject: Re: Refactor "mutually exclusive options" error reporting code in parse_subscription_options