Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-12 Thread Gudmundsson Martin (mg)
Hi all! - checkpoint_segments 1000 - checkpoint_completion_target 0.9 - wal_buffers  256MB - shared_buffers 31 gb - max_connections 500 I see that some of you are using wal_buffers = 256MB. I was under the impression that Postgres will not benefit from higher value than the segment size,

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-12 Thread Luis Antonio Dias de Sá Junior
For me 12000 tps until now 24 core, 150 Gb ram - 5 ssd raid 5 - Debian 7.8 - Postgres 9.3.5 ...with Postgres parameters customized: - checkpoint_segments 1000 - checkpoint_completion_target 0.9 - wal_buffers 256MB - shared_buffers 31 gb - max_connections 500 - effective_io_concurrency 15

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-12 Thread Mark Kirkwood
On 13/02/15 00:20, Gudmundsson Martin (mg) wrote: Hi all! - checkpoint_segments 1000 - checkpoint_completion_target 0.9 - wal_buffers 256MB - shared_buffers 31 gb - max_connections 500 I see that some of you are using wal_buffers = 256MB. I was under the impression that Postgres will not

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-12 Thread Graeme B. Bell
1. O/S Under O/S, don't forget to mention linux kernel version. We saw a MASSIVE increase in TPS (I think it was a doubling? Don't have the data to hand right now) on our multicore RHEL6 servers, when moving from a stock RHEL6 kernel to an ELREPO 3.18 series kernel. That's what 10 years of

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-10 Thread Luis Antonio Dias de Sá Junior
No problem with this. If anyone want to specify more details. But I want to know how far postgres can go. No matter OS or other variables. Gavin, you got more than 12000 TPS? 2015-02-09 19:29 GMT-02:00 Gavin Flower gavinflo...@archidevsys.co.nz: On 10/02/15 08:30, Luis Antonio Dias de Sá

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-10 Thread Graeme B. Bell
I'd suggest you run it on a large ramdisk with fsync turned off on a 32 core computer, see what you get, that will be a good indication of a maximum. Keep in mind though that 'postgres' with fsync (vs. without) is such a different creature that the comparison isn't meaningful. Similarly

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-10 Thread Mark Kirkwood
On 10/02/15 10:29, Gavin Flower wrote: On 10/02/15 08:30, Luis Antonio Dias de Sá Junior wrote: Hi, A survay: with pgbench using TPS-B, what is the maximum TPS you're ever seen? For me: 12000 TPS. -- Luis Antonio Dias de Sá Junior Important to specify: 1. O/S 2. version of PostgreSQL 3.

Re: [PERFORM] Survey: Max TPS you've ever seen

2015-02-09 Thread Gavin Flower
On 10/02/15 08:30, Luis Antonio Dias de Sá Junior wrote: Hi, A survay: with pgbench using TPS-B, what is the maximum TPS you're ever seen? For me: 12000 TPS. -- Luis Antonio Dias de Sá Junior Important to specify: 1. O/S 2. version of PostgreSQL 3. PostgreSQL configuration 4. hardware