pg.ddx.io pgsql-performance@postgresql.org mailing list archive
help / color / mirror / Atom feedStreaming replication performance
3+ messages / 3 participants
[nested] [flat]
* Streaming replication performance
@ 2011-08-11 14:46 Antonin Faltynek <pin007@gmail.com>
0 siblings, 1 reply; 3+ messages in thread
From: Antonin Faltynek @ 2011-08-11 14:46 UTC (permalink / raw)
To: pgsql-performance
Hi all,
I'm testing Streamin replication with one hot standby node and I'm
experiencing high delay of hot standby node.
When I reach aprox. 50 transactions per second where every transaction
includes only simple "UPDATE status SET number = number + 1 WHERE name =
'iterations'", the standby server falls behind master and the delay is several
minutes. In comparison, Slony1 have no issues with the same test and its delay
is only several seconds.
Am I missing some configuration that affects speed of streaming replicaiton?
Both servers are connected using 1Gb network, CPU usage of both servers is
low, disk latency is not issue.
master server postgresql.conf
-----
wal_level = 'hot_standby'
vacuum_defer_cleanup_age = 10000
checkpoint_segments = 8
checkpoint_timeout = 30s
archive_mode = off
max_wal_senders = 3
wal_sender_delay = 10ms
wal_keep_segments = 128
-----
standby server postgresql.conf
-----
hot_standby = on
max_standby_streaming_delay = 30s
-----
standby server recovery.conf
-----
standby_mode = 'on'
primary_conninfo = 'host=master port=5432 user=replication2 password=password'
trigger_file = '/var/lib/postgresql/9.0/main/replica_trigger'
-----
Thank you for any advice
Antonin Faltynek
^ permalink raw reply [nested|flat] 3+ messages in thread
* Re: Streaming replication performance
@ 2011-08-12 16:47 Merlin Moncure <mmoncure@gmail.com>
parent: Antonin Faltynek <pin007@gmail.com>
0 siblings, 1 reply; 3+ messages in thread
From: Merlin Moncure @ 2011-08-12 16:47 UTC (permalink / raw)
To: pin007@gmail.com; +Cc: pgsql-performance
On Thu, Aug 11, 2011 at 9:46 AM, Antonin Faltynek <pin007@gmail.com> wrote:
> Hi all,
>
> I'm testing Streamin replication with one hot standby node and I'm
> experiencing high delay of hot standby node.
>
> When I reach aprox. 50 transactions per second where every transaction
> includes only simple "UPDATE status SET number = number + 1 WHERE name =
> 'iterations'", the standby server falls behind master and the delay is several
> minutes. In comparison, Slony1 have no issues with the same test and its delay
> is only several seconds.
>
> Am I missing some configuration that affects speed of streaming replicaiton?
hm -- how are you determining the standby is behind? are you running
any other queries on the standby?
merlin
^ permalink raw reply [nested|flat] 3+ messages in thread
* Re: Streaming replication performance
@ 2011-08-16 20:03 Pin007 <pin007@gmail.com>
parent: Merlin Moncure <mmoncure@gmail.com>
0 siblings, 0 replies; 3+ messages in thread
From: Pin007 @ 2011-08-16 20:03 UTC (permalink / raw)
To: Merlin Moncure <mmoncure@gmail.com>; +Cc: pgsql-performance
Dne Pá 12. srpna 2011 18:47:24 Merlin Moncure napsal(a):
> On Thu, Aug 11, 2011 at 9:46 AM, Antonin Faltynek <pin007@gmail.com> wrote:
> > Hi all,
> >
> > I'm testing Streamin replication with one hot standby node and I'm
> > experiencing high delay of hot standby node.
> >
> > When I reach aprox. 50 transactions per second where every transaction
> > includes only simple "UPDATE status SET number = number + 1 WHERE name =
> > 'iterations'", the standby server falls behind master and the delay is
> > several minutes. In comparison, Slony1 have no issues with the same test
> > and its delay is only several seconds.
> >
> > Am I missing some configuration that affects speed of streaming
> > replicaiton?
>
> hm -- how are you determining the standby is behind? are you running
> any other queries on the standby?
>
> merlin
Simply, during test I'm generating series of numbers that I'm periodicaly
checking on both master and stand by server.
Also I'm measuring several server metrics, like CPU usage, disk usage, network
bandwidth.
When testing streaming replication network usage does not get over 1Mbps and
catching master takes several tens of minutes, while during test of WAL file
shipping, network usage get to 3-5Mpbs and stand by server does not fall
behind master more than several seconds.
There is no other traffic on stand by server, only periodical check of test
table (SELECT * FROM status) every 2s. Extension of this check interval does
not affect behavior of streaming replication.
Tonda
^ permalink raw reply [nested|flat] 3+ messages in thread
end of thread, other threads:[~2011-08-16 20:03 UTC | newest]
Thread overview: 3+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2011-08-11 14:46 Streaming replication performance Antonin Faltynek <pin007@gmail.com>
2011-08-12 16:47 ` Merlin Moncure <mmoncure@gmail.com>
2011-08-16 20:03 ` Pin007 <pin007@gmail.com>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox