Re: MusicBrainz postgres performance issues

Tomas Vondra

Sunday, March 15, 2015 7:41 PM

On 15.3.2015 23:47, Andres Freund wrote:

On 2015-03-15 12:25:07 -0600, Scott Marlowe wrote:

Here's the problem with a large shared_buffers on a machine that's
getting pushed into swap. It starts to swap BUFFERs. Once buffers
start getting swapped you're not just losing performance, that huge
shared_buffers is now working against you because what you THINK are
buffers in RAM to make things faster are in fact blocks on a hard
drive being swapped in and out during reads. It's the exact opposite
of fast. :)

IMNSHO that's tackling things from the wrong end. If 12GB of shared 
buffers drive your 48GB dedicated OLTP postgres server into swapping
out actively used pages, the problem isn't the 12GB of shared
buffers, but that you require so much memory for other things. That
needs to be fixed.

I second this opinion.

As was already pointed out, the 500 connections is rather insane
(assuming the machine does not have hundreds of cores).

If there are memory pressure issues, it's likely because many queries
are performing memory-expensive operations at the same time (might even
be a bad estimate causing hashagg to use much more than work_mem).

But! We haven't even established that swapping is an actual problem
here. The ~2GB of swapped out memory could just as well be the java raid
controller management monstrosity or something similar. Those pages
won't ever be used and thus can better be used to buffer IO.

You can check what's actually swapped out using:
grep ^VmSwap /proc/[0-9]*/status|grep -v '0 kB'

For swapping to be actually harmful you need to have pages that are 
regularly swapped in. vmstat will tell.

I've already asked for vmstat logs, so let's wait.

In a concurrent OLTP workload (~450 established connections do
suggest that) with a fair amount of data keeping the hot data set in 
shared_buffers can significantly reduce problems. Constantly
searching for victim buffers isn't a nice thing, and that will happen
if your most frequently used data doesn't fit into s_b. On the other
hand, if your data set is so large that even the hottest part doesn't
fit into memory (perhaps because there's no hottest part as there's
no locality at all), a smaller shared buffers can make things more
efficient, because the search for replacement buffers is cheaper with
a smaller shared buffers setting.

I've met many systems with max_connections values this high, and it was
mostly idle connections because of separate connection pools on each
application server. So mostly idle (90% of the time), but at peak time
all the application servers want to od stuff at the same time. And it
all goes KABOOOM! just like here.

Andres Freund

Sunday, March 15, 2015 6:47 PM

IMNSHO that's tackling things from the wrong end. If 12GB of shared
buffers drive your 48GB dedicated OLTP postgres server into swapping out
actively used pages, the problem isn't the 12GB of shared buffers, but
that you require so much memory for other things. That needs to be
fixed.

But! We haven't even established that swapping is an actual problem
here. The ~2GB of swapped out memory could just as well be the java raid
controller management monstrosity or something similar. Those pages
won't ever be used and thus can better be used to buffer IO.

You can check what's actually swapped out using:
grep ^VmSwap /proc/[0-9]*/status|grep -v '0 kB'

For swapping to be actually harmful you need to have pages that are
regularly swapped in. vmstat will tell.

In a concurrent OLTP workload (~450 established connections do suggest
that) with a fair amount of data keeping the hot data set in
shared_buffers can significantly reduce problems. Constantly searching
for victim buffers isn't a nice thing, and that will happen if your most
frequently used data doesn't fit into s_b. On the other hand, if your
data set is so large that even the hottest part doesn't fit into memory
(perhaps because there's no hottest part as there's no locality at all),
a smaller shared buffers can make things more efficient, because the
search for replacement buffers is cheaper with a smaller shared buffers
setting.

Greetings,

Andres Freund

Scott Marlowe

Sunday, March 15, 2015 2:25 PM

Here's the problem with a large shared_buffers on a machine that's
getting pushed into swap. It starts to swap BUFFERs. Once buffers
start getting swapped you're not just losing performance, that huge
shared_buffers is now working against you because what you THINK are
buffers in RAM to make things faster are in fact blocks on a hard
drive being swapped in and out during reads. It's the exact opposite
of fast. :)

Andres Freund

Sunday, March 15, 2015 1:46 PM

That imo doesn't really have anything to do with it. The primary benefit
of a BBU with writeback caching is accelerating (near-)synchronous
writes. Like the WAL. But, besides influencing the default for
wal_buffers, a larger shared_buffers doesn't change the amount of
synchronous writes.

Greetings,

Andres Freund

Ilya Kosmodemiansky

Sunday, March 15, 2015 1:42 PM

On Sun, Mar 15, 2015 at 8:20 PM, Andres Freund <andres@2ndquadrant.com> wrote:

On 2015-03-15 11:09:34 -0600, Scott Marlowe wrote:

shared_mem of 12G is almost always too large. I'd drop it down to ~1G or so.

I think that's a outdated wisdom, i.e. not generally true.

Quite agreed. With note, that proper configured controller with BBU is needed.

A new enough kernel, a sane filesystem
(i.e. not ext3) and sane checkpoint configuration takes care of most of
the other disadvantages.

Most likely. And better to be sure that filesystem mounted without barrier.

And I agree with Scott - 64MB work mem AND max_connections = 500 is a
recipe for disaster. The problem could be in session mode of
pgbouncer. If you can work with transaction mode - do it.


Best regards,
Ilya Kosmodemiansky,

PostgreSQL-Consulting.com
tel. +14084142500
cell. +4915144336040
ik@postgresql-consulting.com