MIME-Version: 1.0
References: 
 <CAJk5AtYBpy2ROKzv+se+LXpM4PZdzifnY_rZvamjHpBELxYUog@mail.gmail.com>
In-Reply-To: 
 <CAJk5AtYBpy2ROKzv+se+LXpM4PZdzifnY_rZvamjHpBELxYUog@mail.gmail.com>
From: Jeffery Janes <jeff.janes@gmail.com>
Date: Sun, 21 Jun 2026 15:19:14 -0400
Message-ID: 
 <CAMkU=1xvG99AY7-SiOjppHVZXGdFYzRHOD1GFeSV_mBj84t53w@mail.gmail.com>
Subject: Re: TPS in direct Postgres Vs TPS in pgbouncer
To: Raj <rajeshkumar.dba09@gmail.com>
Cc: Pgsql-admin <pgsql-admin@lists.postgresql.org>
Content-Type: multipart/alternative; boundary="0000000000006f7d7c0654c86738"
Archived-At: 
 <https://www.postgresql.org/message-id/CAMkU%3D1xvG99AY7-SiOjppHVZXGdFYzRHOD1GFeSV_mBj84t53w%40mail.gmail.com>
Precedence: bulk

--0000000000006f7d7c0654c86738
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sun, Jun 21, 2026 at 9:24=E2=80=AFAM Raj <rajeshkumar.dba09@gmail.com> w=
rote:

> Hi All,
>
> When the application team is using posytgres directly for 17.9 version,
> they get necessary TPS.
>
> When they use pgbouncer, they get 50% reduction in TPS.
>
> What are general advices?
>

What is the transaction being used?  If it is something very simple, like
pgbench's built-in -S, then I think this is expected.  Injecting pgbouncer
doubles the amount of network/IPC thrashing needed.  If you have a meaty
transaction, this might not matter but on a very light transaction you
might be doubling something which is already counting for 90% of the time.

Cheers,

Jeff

--0000000000006f7d7c0654c86738
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote g=
mail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Sun, Jun 21,=
 2026 at 9:24=E2=80=AFAM Raj &lt;<a href=3D"mailto:rajeshkumar.dba09@gmail.=
com">rajeshkumar.dba09@gmail.com</a>&gt; wrote:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex"><div dir=3D"auto">Hi All,<div dir=3D"auto"=
><br></div><div dir=3D"auto">When the application team is using posytgres d=
irectly for 17.9 version, they get necessary TPS.</div><div dir=3D"auto"><b=
r></div><div dir=3D"auto">When they use pgbouncer, they get 50% reduction i=
n TPS.</div><div dir=3D"auto"><br></div><div dir=3D"auto">What are general =
advices?</div></div></blockquote><div><br></div><div>What is the transactio=
n being used?=C2=A0 If it is something very simple, like pgbench&#39;s=C2=
=A0built-in -S, then I think this is expected.=C2=A0 Injecting pgbouncer do=
ubles the amount of network/IPC thrashing needed.=C2=A0 If you have a meaty=
 transaction, this might not matter but on a very light transaction you mig=
ht be doubling something which is already counting for 90% of the time.</di=
v><div><br></div><div>Cheers,</div><div><br></div><div>Jeff</div></div></di=
v>

--0000000000006f7d7c0654c86738--