Horton

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

John Naylor <john.naylor@enterprisedb.com>

View thread View on postgr.es
From: John Naylor <john.naylor@enterprisedb.com>
To: Masahiko Sawada <sawada.mshk@gmail.com>
Cc: Nathan Bossart <nathandbossart@gmail.com>, Andres Freund <andres@anarazel.de>, Matthias van de Meent <boekewurm+postgres@gmail.com>, Yura Sokolov <y.sokolov@postgrespro.ru>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2022-11-21T09:30:29Z
Lists: pgsql-hackers 
On Mon, Nov 21, 2022 at 3:43 PM Masahiko Sawada <sawada.mshk@gmail.com>
wrote:
>
> On Mon, Nov 21, 2022 at 4:20 PM John Naylor
> <john.naylor@enterprisedb.com> wrote:

> > Assuming the smallest node is fixed size (i.e. fanout/capacity member
not part of the common set, so only part of variable-sized nodes), 3 has a
nice property: no wasted padding space:
> >
> > node4: 5 + 4+(7) + 4*8 = 48 bytes
> > node3: 5 + 3     + 3*8 = 32
>
> IIUC if we store the fanout member only in variable-sized nodes,
> rt_node has only count, shift, and chunk, so 4 bytes in total. If so,
> the size of node3 (ie. fixed-sized node) is (4 + 3 + (1) + 3*8)? The
> size doesn't change but there is 1 byte padding space.

I forgot to mention I'm assuming no pointer-tagging for this exercise.
You've demonstrated it can be done in a small amount of code, and I hope we
can demonstrate a speedup in search. Just in case there is some issue with
portability, valgrind, or some other obstacle, I'm being pessimistic in my
calculations.

> Also, even if we have the node3 a variable-sized node, size class 1
> for node3 could be a good choice since it also doesn't need padding
> space and could be a good alternative to path compression.
>
> node3         :  5 + 3 + 3*8 = 32 bytes
> size class 1 : 5 + 3 + 1*8 = 16 bytes

Precisely! I have that scenario in my notes as well -- it's quite
compelling.

--
John Naylor
EDB: http://www.enterprisedb.com

Commits

  1. radixtree: Fix SIGSEGV at update of embeddable value to non-embeddable.

  2. Get rid of anonymous struct

  3. Teach radix tree to embed values at runtime

  4. Teach TID store to skip bitmap for small numbers of offsets

  5. Use bump context for TID bitmaps stored by vacuum

  6. Fix alignment of stack variable

  7. Use TidStore for dead tuple TIDs storage during lazy vacuum.

  8. Rethink create and attach APIs of shared TidStore.

  9. Fix inconsistent function prototypes with function definitions.

  10. Fix a calculation in TidStoreCreate().

  11. Fix potential integer handling issue in radixtree.h.

  12. Add TIDStore, to store sets of TIDs (ItemPointerData) efficiently.

  13. Fix link error for test_radixtree module on Windows

  14. Blind attempt to fix ODR violations

  15. Fix incorrect format specifier for int64

  16. Fix redefinition of typedefs

  17. Add template for adaptive radix tree

  18. Fix signedness error in 9f225e992 for gcc

  19. Introduce helper SIMD functions for small byte arrays

  20. Optimize vacuuming of relations with no indexes.

  21. Add bound check before bsearch() for performance

  22. Allocate consecutive blocks during parallel seqscans