Re: Survey: Max TPS you've ever seen - Mailing list pgsql-performance

From Graeme B. Bell
Subject Re: Survey: Max TPS you've ever seen
Date
Msg-id 8D72ADF3-B9B5-4A98-9F05-F7821438593E@skogoglandskap.no
Whole thread Raw
In response to Re: Survey: Max TPS you've ever seen  (Luis Antonio Dias de Sá Junior <luisjunior.sa@gmail.com>)
List pgsql-performance
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
goodindication 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 'postgres' on volatile backing store vs. non-volatile isn't really a meaningful comparison.

There's also a question here about the 't' in TPS. If you have no fsync and volatile storage, are you really doing
'transactions'?Depending on the definition you take, a transaction may have some sense of 'reliability' or atomicity
whichisn't reflected well in a ramdisk/no-fsync benchmark.  

It's probably not ideal to fill a mailing list with numbers that have no meaning attached to them, so why not set up a
littleweb database or Google doc to record max TPS and how it was achieved? 

For example, imagine I tell you that the highest I've achieved is 1240000 tps. How does it help you if I say that?

Graeme Bell

On 10 Feb 2015, at 11:48, Luis Antonio Dias de Sá Junior <luisjunior.sa@gmail.com> wrote:

> 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 <GavinFlower@archidevsys.co.nz>:
> 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 configuration
> 5. anything else that might affect performance
>
> I suspect that Linux will out perform Microsoft on the same hardware, and optimum configuration for both O/S's...
>
>
> Cheers,
> Gavin
>
>
> --
> Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-performance
>
>
>
> --
> Luis Antonio Dias de Sá Junior



pgsql-performance by date:

Previous
From: Luis Antonio Dias de Sá Junior
Date:
Subject: Re: Survey: Max TPS you've ever seen
Next
From: Mark Kirkwood
Date:
Subject: Re: Survey: Max TPS you've ever seen