pgsql: Fix VACUUM so that it always updates pg_class.reltuples/relpages

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: pgsql-committers(at)postgresql(dot)org
Subject: pgsql: Fix VACUUM so that it always updates pg_class.reltuples/relpages
Date: 2011-05-30 21:08:00
Message-ID: E1QR9gu-0005AW-7a@gemulon.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-committers

Fix VACUUM so that it always updates pg_class.reltuples/relpages.

When we added the ability for vacuum to skip heap pages by consulting the
visibility map, we made it just not update the reltuples/relpages
statistics if it skipped any pages. But this could leave us with extremely
out-of-date stats for a table that contains any unchanging areas,
especially for TOAST tables which never get processed by ANALYZE. In
particular this could result in autovacuum making poor decisions about when
to process the table, as in recent report from Florian Helmberger. And in
general it's a bad idea to not update the stats at all. Instead, use the
previous values of reltuples/relpages as an estimate of the tuple density
in unvisited pages. This approach results in a "moving average" estimate
of reltuples, which should converge to the correct value over multiple
VACUUM and ANALYZE cycles even when individual measurements aren't very
good.

This new method for updating reltuples is used by both VACUUM and ANALYZE,
with the result that we no longer need the grotty interconnections that
caused ANALYZE to not update the stats depending on what had happened
in the parent VACUUM command.

Also, fix the logic for skipping all-visible pages during VACUUM so that it
looks ahead rather than behind to decide what to do, as per a suggestion
from Greg Stark. This eliminates useless scanning of all-visible pages at
the start of the relation or just after a not-all-visible page. In
particular, the first few pages of the relation will not be invariably
included in the scanned pages, which seems to help in not overweighting
them in the reltuples estimate.

Back-patch to 8.4, where the visibility map was introduced.

Branch
------
REL9_0_STABLE

Details
-------
http://git.postgresql.org/pg/commitdiff/73bd34c81ebd2208f4bd207b83f8a4df8aec82d2

Modified Files
--------------
src/backend/commands/analyze.c | 59 +++++--------
src/backend/commands/vacuum.c | 98 ++++++++++++++++++----
src/backend/commands/vacuumlazy.c | 169 ++++++++++++++++++++++++-------------
src/backend/postmaster/pgstat.c | 17 +---
src/include/commands/vacuum.h | 8 ++-
src/include/pgstat.h | 6 +-
6 files changed, 228 insertions(+), 129 deletions(-)

Browse pgsql-committers by date

  From Date Subject
Next Message Tom Lane 2011-05-30 23:16:58 pgsql: Fix portability bugs in use of credentials control messages for
Previous Message Peter Eisentraut 2011-05-30 20:36:49 pgsql: Suppress foreign data wrappers and foreign servers in partial du