Re: Database block lifecycle

From: pinker <pinker(at)onet(dot)eu>
To: John R Pierce <pierce(at)hogranch(dot)com>, pgsql-general(at)postgresql(dot)org
Subject: Re: Database block lifecycle
Date: 2014-08-12 22:29:36
Message-ID: 53EA9550.5030503@onet.eu
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

yes, I know the count is quite high. It is the max value we've
estimated, but probably on average day it will be 100-200, and yes we
use pgpool.

Am 13.08.2014 00:09, schrieb John R Pierce:
> On 8/12/2014 2:41 PM, pinker wrote:
>> btw. 512MB if we assume up to 600 connection is a reasonable value?
>
> thats an insanely high connection count, if you actually expect those
> connections to be executing concurrent queries, unless you have
> something north of 100 CPU cores.
>
> you'd be much better to have a MUCH smaller connection count, and use
> a connection pooler such as pgbouncer, in transaction mode... let 600
> client htreads connect to the pooler, but have the pooler share maybe
> 4X your CPU core/thread count of actual connections for transactions
> in progress.
>
>
>
>
>

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message John R Pierce 2014-08-12 22:39:35 Re: Database block lifecycle
Previous Message John R Pierce 2014-08-12 22:09:35 Re: Database block lifecycle