pgsql: Reduce ProcArrayLock contention by removing backends in batches. - Mailing list pgsql-committers

From Robert Haas
Subject pgsql: Reduce ProcArrayLock contention by removing backends in batches.
Date
Msg-id E1ZNNg0-0003hW-Kz@gemulon.postgresql.org
Whole thread Raw
List pgsql-committers
Reduce ProcArrayLock contention by removing backends in batches.

When a write transaction commits, it must clear its XID advertised via
the ProcArray, which requires that we hold ProcArrayLock in exclusive
mode in order to prevent concurrent processes running GetSnapshotData
from seeing inconsistent results.  When many processes try to commit
at once, ProcArrayLock must change hands repeatedly, with each
concurrent process trying to commit waking up to acquire the lock in
turn.  To make things more efficient, when more than one backend is
trying to commit a write transaction at the same time, have just one
of them acquire ProcArrayLock in exclusive mode and clear the XIDs of
all processes in the group.  Benchmarking reveals that this is much
more efficient at very high client counts.

Amit Kapila, heavily revised by me, with some review also from Pavan
Deolasee.

Branch
------
master

Details
-------
http://git.postgresql.org/pg/commitdiff/0e141c0fbb211bdd23783afa731e3eef95c9ad7a

Modified Files
--------------
src/backend/access/transam/README   |    3 +
src/backend/storage/ipc/procarray.c |  166 ++++++++++++++++++++++++++++++-----
src/backend/storage/lmgr/proc.c     |    5 ++
src/include/storage/proc.h          |   12 +++
4 files changed, 166 insertions(+), 20 deletions(-)


pgsql-committers by date:

Previous
From: Kevin Grittner
Date:
Subject: pgsql: Fix `make installcheck` for serializable transactions.
Next
From: Robert Haas
Date:
Subject: pgsql: Fix incorrect calculation in shm_mq_receive.