Re: [PoC] Improve dead tuple storage for lazy vacuum

From: John Naylor <johncnaylorls(at)gmail(dot)com>
To: Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, Nathan Bossart <nathandbossart(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, Matthias van de Meent <boekewurm+postgres(at)gmail(dot)com>, Yura Sokolov <y(dot)sokolov(at)postgrespro(dot)ru>
Subject: Re: [PoC] Improve dead tuple storage for lazy vacuum
Date: 2024-04-07 09:44:56
Message-ID: CANWCAZZQFfxvzO8yZHFWtQV+Z2gAMv1ku16Vu7KWmb5kZQyd1w@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sun, Apr 7, 2024 at 9:08 AM John Naylor <johncnaylorls(at)gmail(dot)com> wrote:
> I've attached a mostly-polished update on runtime embeddable values,
> storing up to 3 offsets in the child pointer (1 on 32-bit platforms).

And...since there's a new bump context patch, I wanted to anticipate
squeezing an update on top of that, if that gets committed. 0004/5 are
the v6 bump context, and 0006 uses it for vacuum. The rest are to show
it works -- the expected.out changes make possible problems in CI
easier to see. The allocation size is 16 bytes, so this difference is
entirely due to lack of chunk header:

aset: 6619136
bump: 5047296

(Note: assert builds still have the chunk header for sanity checking,
so this was done in a more optimized build)

Attachment Content-Type Size
v84-0005-Introduce-a-bump-memory-allocator.patch text/x-patch 30.2 KB
v84-0008-DEV-compare-bump-context-in-tests.patch text/x-patch 889 bytes
v84-0004-Enlarge-bit-space-for-MemoryContextMethodID.patch text/x-patch 12.1 KB
v84-0007-DEV-log-memory-usage-in-tests.patch text/x-patch 2.6 KB
v84-0006-Use-bump-context-for-vacuum-s-TID-storage.patch text/x-patch 5.5 KB
v84-0003-Teach-radix-tree-to-embed-values-at-runtime.patch text/x-patch 8.1 KB
v84-0002-pgindent.patch text/x-patch 4.2 KB
v84-0001-store-offsets-in-the-header.patch text/x-patch 5.5 KB

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Jelte Fennema-Nio 2024-04-07 10:04:48 Re: Flushing large data immediately in pqcomm
Previous Message Daniel Gustafsson 2024-04-07 09:39:39 Re: Cluster::restart dumping logs when stop fails