Re: Tuning Postgres for Single connection use

On 04/14/2014 05:46 PM, Nick Eubank wrote:
>
> Any rules of thumb for |work_mem|, |maintenance_work_mem|,
> |shared_buffer|, etc. for a database that DOESN'T anticipate
> concurrent connections and that is doing lots of aggregate functions
> on large tables? All the advice I can find online on tuning (this
> <http://wiki.postgresql.org/wiki/Performance_Optimization>, this
> <http://media.revsys.com/talks/djangocon/2011/secrets-of-postgresql-performance.pdf>,
> this
> <http://www.revsys.com/writings/postgresql-performance.html> etc.) is
> written for people anticipating lots of concurrent connections.
>
> I'm a social scientist looking to use Postgres not as a database to be
> shared by multiple users, but rather as my own tool for manipulating a
> massive data set (I have 5 billion transaction records (600gb in csv)
> and want to pull out unique user pairs, estimate aggregates for
> individual users, etc.). This also means almost no writing, except to
> creation of new tables based on selections from the main table.
>
> I'm on a Windows 8 VM with 16gb ram, SCSI VMware HD, and 3 cores if
> that's important.
>
>


First up would probably be "don't run on Windows". shared_buffers above
512Mb causes performance to degrade on Windows, while that threshold is
much higher on *nix.

cheers

andrew