Horton

Re: Emit fewer vacuum records by reaping removable tuples during pruning

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: Melanie Plageman <melanieplageman@gmail.com>
Cc: Robert Haas <robertmhaas@gmail.com>, Michael Paquier <michael@paquier.xyz>, Pg Hackers <pgsql-hackers@postgresql.org>, Andres Freund <andres@anarazel.de>
Date: 2024-01-12T20:01:59Z
Lists: pgsql-hackers 
On Fri, Jan 12, 2024 at 1:52 PM Melanie Plageman
<melanieplageman@gmail.com> wrote:
> On Fri, Jan 12, 2024 at 1:07 PM Peter Geoghegan <pg@bowt.ie> wrote:
> > What is "space_freed"? Isn't that something from your uncommitted patch?
>
> Yes, I was mixing the two together.

An understandable mistake.

> I just want to make sure that we agree that, on master, when
> lazy_scan_prune() is called, the logic for whether or not to update
> the FSM after the first pass is:
>
> indexes == 0 || !has_lpdead_items || !index_vacuuming
>
> and when lazy_scan_noprune() is called, the logic for whether or not
> to update the FSM after the first pass is:
>
> indexes == 0 || !has_lpdead_items
>
> Those seem different to me.

Right. As I said to Robert just now, I can now see that they're
slightly different conditions.

FWIW my brain was just ignoring " || !index_vacuuming". I dismissed it
as an edge-case, only relevant when the failsafe has kicked in. Which
it is. But that's still no reason to allow an inconsistency that we
can easily just avoid.

-- 
Peter Geoghegan

Commits

  1. Combine FSM updates for prune and no-prune cases.

  2. Remove LVPagePruneState.

  3. Move VM update code from lazy_scan_heap() to lazy_scan_prune().

  4. Optimize vacuuming of relations with no indexes.

  5. Be more consistent about whether to update the FSM while vacuuming.

  6. Remove hastup from LVPagePruneState.

  7. Use scanned_pages to decide when to failsafe check.

  8. Simplify lazy_scan_heap's handling of scanned pages.

  9. While vacuuming a large table, update upper-level FSM data every so often.