Horton

Re: [HACKERS] Custom compression methods

Dilip Kumar <dilipbalaut@gmail.com>

View thread View on postgr.es
From: Dilip Kumar <dilipbalaut@gmail.com>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Justin Pryzby <pryzby@telsasoft.com>, Tomas Vondra <tomas.vondra@2ndquadrant.com>, Alexander Korotkov <a.korotkov@postgrespro.ru>, David Steele <david@pgmasters.net>, Ildus Kurbangaliev <i.kurbangaliev@gmail.com>, Dmitry Dolgov <9erthalion6@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2021-03-16T13:05:21Z
Lists: pgsql-hackers 
On Tue, Mar 16, 2021 at 4:07 PM Dilip Kumar <dilipbalaut@gmail.com> wrote:
>
> INSERT TIME
> Head: 17418.299 ms Patch: 20956.231 ms
>
> CTAS TIME:
> Head: 12837.872 ms Patch: 16775.739 ms

On quick analysis with perf it appeared that the performance is
degrading because of deforming

-   16.19%     3.54%  postgres  postgres            [.]
CompareCompressionMethodAndDecompress
   - 12.65% CompareCompressionMethodAndDecompress
      - 12.57% slot_getallattrs
         - 12.56% slot_getsomeattrs
            - 12.53% slot_getsomeattrs_int
               - 12.50% tts_buffer_heap_getsomeattrs
                    slot_deform_heap_tuple

So I think in the case of direct insert it needs to deform because I
am calling CompareCompressionMethodAndDecompress after ExecCopySlot
and that is why we have to deform every time so maybe that can be
avoided by calling CompareCompressionMethodAndDecompress before
ExecCopySlot.  But in the case of CTAS or INSERT INTO SELECT we can
not avoid deforming because we might get the formed tuple from the
source table.  I put a temporary hack to keep the map of the varlena
attribute and use it across the tuple but it did not improve the
performance in this case because the main bottleneck is
slot_getallattrs.   I think this should help where we don't have any
varlena, but first I need to test whether we can any performance
regression in those cases at all.

--
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com

Commits

  1. docs: Update TOAST storage docs for configurable compression.

  2. Further tweaking of pg_dump's handling of default_toast_compression.

  3. Fix interaction of TOAST compression with expression indexes.

  4. Tidy up more loose ends related to configurable TOAST compression.

  5. Short-circuit slice requests that are for more than the object's size.

  6. Mostly-cosmetic adjustments of TOAST-related macros.

  7. Remove useless configure probe for <lz4/lz4.h>.

  8. Error on invalid TOAST compression in CREATE or ALTER TABLE.

  9. docs: Fix omissions related to configurable TOAST compression.

  10. More code cleanup for configurable TOAST compression.

  11. Bring configure support for LZ4 up to snuff.

  12. Make compression.sql regression test independent of default.

  13. Use valid compression method in brin_form_tuple

  14. Fix up pg_dump's handling of per-attribute compression options.

  15. Allow configurable LZ4 TOAST compression.

  16. Fix inconsistencies in the code

  17. Mostly-cosmetic improvements in memory chunk header alignment coding.

  18. Allow numeric to use a more compact, 2-byte header in many cases.