Re: Auto-tuning work_mem and maintenance_work_mem

From: Josh Berkus <josh(at)agliodbs(dot)com>
To: Merlin Moncure <mmoncure(at)gmail(dot)com>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, Andrew Dunstan <andrew(at)dunslane(dot)net>, Bruce Momjian <bruce(at)momjian(dot)us>, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com>, Andres Freund <andres(at)2ndquadrant(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Auto-tuning work_mem and maintenance_work_mem
Date: 2013-10-18 16:14:48
Message-ID: 52615E78.8000600@agliodbs.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

All,

So, I did an informal survey last night a SFPUG, among about 30
PostgreSQL DBAs and developers. While hardly a scientific sample, it's
a data point on what we're looking at for servers.

Out of the 30, 6 had one or more production instances of PostgreSQL
running on machines or VMs with less than 1GB of RAM. Out of those 5
had already edited their PostgreSQL.conf extensively. Perhaps more
importantly, for four out of the 6, the low-memory Postgres instance(s)
was an older version (8.2 to 9.0) which they did not expect to upgrade.
Also, note that a couple of the 6 were consultants, so they were
speaking for dozens of customer servers.

As a second data point, Christophe and I did a quick survey of the
database of server information on our clients, which include a bunch of
cloud-hosted web companies. We found two PostgreSQL VMs which did not
have 1GB or more RAM, out of a few hundred.

Now, obviously, there's some significant sample bias in the above, but I
think it gives support to the assertion that we shouldn't really be
worrying about PostgresQL running well out-of-the-box on machines with <
1GB of RAM.

--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Josh Berkus 2013-10-18 16:19:10 Re: [PATCH] pg_sleep(interval)
Previous Message Tomas Vondra 2013-10-18 16:06:04 Re: fdw_private and (List*) handling in FDW API