MIME-Version: 1.0
References: 
 <CALj2ACVfF2Uj9NoFy-5m98HNtjHpuD17EDE9twVeJng-jTAe7A@mail.gmail.com>
 <CAPsk3_CzLbMe-D07H5Vo6yWFvyXHh5is7AoPUCFcztrUmf1haw@mail.gmail.com>
 <CALj2ACVzL4uU=hxFpSfkqP4HeFCPucbBTEg6HNf_MPTYm52pHg@mail.gmail.com>
 <CAMm1aWYa1fGKcuG69xGJPNXLQ_9zHrPqhr-ZGdj4so6Exq66MQ@mail.gmail.com>
 <CALj2ACXa-2eEHHaNRwjcF1k9rtH=EJrWvbGJkucdSOD3zP-OUw@mail.gmail.com>
 <CAPsk3_A7079UtVqm2WXXiwadGJ7DucpenmLwnXZgDgXee703Rw@mail.gmail.com>
 <CALj2ACX+LKR7=3TkP83_9cdcXZd+9zhXWokjXyh5tTSi25+ogw@mail.gmail.com>
 <fa2e932eeff472250e2dbacb49d8c43ad282fea9.camel@j-davis.com>
 <CALj2ACW4BvWL_PyvS-ZF5Z70bymLPJWLHXVLbGCZoAwjA6EzeA@mail.gmail.com>
 <783426a05a756b29b5b1ef4d5640f22d999fdec8.camel@j-davis.com>
 <CALj2ACWn0nS=ejzHwdnnmWhRtWT6=+NAMbcMB5w37sCw97_zBA@mail.gmail.com>
In-Reply-To: 
 <CALj2ACWn0nS=ejzHwdnnmWhRtWT6=+NAMbcMB5w37sCw97_zBA@mail.gmail.com>
From: Bharath Rupireddy <bharath.rupireddyforpostgres@gmail.com>
Date: Fri, 20 Mar 2026 09:00:00 -0700
Message-ID: 
 <CALj2ACUyvZ=MF82MEeJOyAFHOxFJ02jgGg4gbUCb3iemMgn4cA@mail.gmail.com>
Subject: Re: Use WALReadFromBuffers in more places
To: Jeff Davis <pgsql@j-davis.com>
Cc: Jingtang Zhang <mrdrivingduck@gmail.com>,
 pgsql-hackers@lists.postgresql.org,
	Nitin Jadhav <nitinjadhavpostgres@gmail.com>
Content-Type: multipart/mixed; boundary="000000000000d3f706064d76ca22"
Archived-At: 
 <https://www.postgresql.org/message-id/CALj2ACUyvZ%3DMF82MEeJOyAFHOxFJ02jgGg4gbUCb3iemMgn4cA%40mail.gmail.com>
Precedence: bulk

--000000000000d3f706064d76ca22
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Sep 24, 2025 at 12:32=E2=80=AFPM Bharath Rupireddy
<bharath.rupireddyforpostgres@gmail.com> wrote:
>
> > On Wed, 2025-09-24 at 07:26 -0700, Bharath Rupireddy wrote:
> > > Right. Reading unflushed WAL buffers for replication was one of the
> > > motivations. But, in general, WALReadFromBuffers has more benefits
> > > since it lets WAL buffers act as a cache for reads, avoiding the need
> > > to re-read WAL from disk for (both physical and logical) replication.
> > > For example, it makes the use of direct I/O for WAL more realistic
> > > and
> > > can provide significant performance benefits [1].
>
> Thanks for looking into this. I did performance analysis with WAL directo=
 I/O to see how reading from WAL buffers affects walsenders: https://www.po=
stgresql.org/message-id/CALj2ACV6rS%2B7iZx5%2BoAvyXJaN4AG-djAQeM1mrM%3DYSDk=
VrUs7g%40mail.gmail.com. Following is from that thread. Please let me know =
if you have any specific cases in mind. I'm happy to run the same test for =
logical replication.
>
> It helps WAL DIO; since there's no OS
> page cache, using WAL buffers as read cache helps a lot. It is clearly
> evident from my experiment with WAL DIO patch [1], see the results [2]
> and attached graph. As expected, WAL DIO brings down the TPS, whereas
> WAL buffers read i.e. this patch brings it up.
>
> [2] Test case is an insert pgbench workload.
> clients HEAD | WAL DIO | WAL DIO & WAL BUFFERS READ | WAL BUFFERS READ
> 1 1404 1070 1424 1375
> 2 1487 796 1454 1517
> 4 3064 1743 3011 3019
> 8 6114 3556 6026 5954
> 16 11560 7051 12216 12132
> 32 23181 13079 23449 23561
> 64 43607 26983 43997 45636
> 128 80723 45169 81515 81911
> 256 110925 90185 107332 114046
> 512 119354 109817 110287 117506
> 768 112435 105795 106853 111605
> 1024 107554 105541 105942 109370
> 2048 88552 79024 80699 90555
> 4096 61323 54814 58704 61743

Thank you all for reviewing this. Please find the attached rebased
patch for further review.

--
Bharath Rupireddy
Amazon Web Services: https://aws.amazon.com

--000000000000d3f706064d76ca22
Content-Type: application/x-patch;
	name="v4-0001-Use-WALReadFromBuffers-in-more-places.patch"
Content-Disposition: attachment;
	filename="v4-0001-Use-WALReadFromBuffers-in-more-places.patch"
Content-Transfer-Encoding: base64
Content-ID: <f_mmyjcqck0>
X-Attachment-Id: f_mmyjcqck0
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--000000000000d3f706064d76ca22--