Hi,
I've previously posted a patch at
http://archives.postgresql.org/message-id/20141010160020.GG6670%40alap3.anarazel.de
that reduces contention in StrategyGetBuffer() by making the clock sweep
lockless. Robert asked me to post it to a new thread; I originally
wrote it to see some other contention in more detail, that's why it
ended up in that thread...
The performance numbers I've quoted over there are:
> Test:
> pgbench -M prepared -P 5 -S -c 496 -j 496 -T 5000
> on a scale=1000 database, with 4GB of shared buffers.
>
> Before:
> progress: 40.0 s, 136252.3 tps, lat 3.628 ms stddev 4.547
> progress: 45.0 s, 135049.0 tps, lat 3.660 ms stddev 4.515
> progress: 50.0 s, 135788.9 tps, lat 3.640 ms stddev 4.398
> progress: 55.0 s, 135268.4 tps, lat 3.654 ms stddev 4.469
> progress: 60.0 s, 134991.6 tps, lat 3.661 ms stddev 4.739
>
> after:
> progress: 40.0 s, 207701.1 tps, lat 2.382 ms stddev 3.018
> progress: 45.0 s, 208022.4 tps, lat 2.377 ms stddev 2.902
> progress: 50.0 s, 209187.1 tps, lat 2.364 ms stddev 2.970
> progress: 55.0 s, 206462.7 tps, lat 2.396 ms stddev 2.871
> progress: 60.0 s, 210263.8 tps, lat 2.351 ms stddev 2.914
Imo the patch doesn't complicate the logic noticeably...
I do wonder if we could make the freelist accesses lockless as well -
but I think that's harder. So I don't want to mix that with this.
Greetings,
Andres Freund
--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
