From: Tom Lane <tgl@sss.pgh.pa.us>
To: Craig James <cjames@emolecules.com>
cc: "pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Subject: Re: Are many idle connections bad?
In-reply-to: 
 <CAFwQ8rcnqy4F4zDg+UX0A5dLbJ=4ifCDk7WN21y7w5HLtYxo7w@mail.gmail.com>
References: 
 <CAFwQ8rcnqy4F4zDg+UX0A5dLbJ=4ifCDk7WN21y7w5HLtYxo7w@mail.gmail.com>
Comments: In-reply-to Craig James <cjames@emolecules.com>
 message dated "Sat, 25 Jul 2015 07:50:35 -0700"
Date: Sat, 25 Jul 2015 11:04:51 -0400
Message-ID: <8824.1437836691@sss.pgh.pa.us>
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

Craig James <cjames@emolecules.com> writes:
> ... This would result in a thousand
> or so Postgres connections on a machine with 32 CPUs.

> So the question is: do idle connections impact performance?

Yes.  Those connections have to be examined when gathering snapshot
information, since you don't know that they're idle until you look.
So the cost of taking a snapshot is proportional to the total number
of connections, even when most are idle.  This sort of situation
is known to aggravate contention for the ProcArrayLock, which is a
performance bottleneck if you've got lots of CPUs.

You'd be a lot better off with a pooler.

(There has been, and continues to be, interest in getting rid of this
bottleneck ... but it's a problem in all existing Postgres versions.)

			regards, tom lane


-- 
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance