Re: postgresql 8.3 tps rate

On Mon, 26 Jan 2009, M. Edward (Ed) Borasky wrote:

> Is there a howto somewhere on disabling this on a Seagate Barracuda?

http://inferno.slug.org/cgi-bin/wiki?Western_Digital_NCQ is a good
discussion of disabling NCQ support under Linux (both in user-space and
directly in the kernel itself). A good summary of the problems with the
WD NCQ support alluded to there are at http://lkml.org/lkml/2007/4/3/159
if you're curious.

It's a tough time to be picking up inexpensive consumer SATA disks right
now. Seagate's drive reliability has been falling hard the last couple of
years, but all the WD drives I've started trying out instead have just
awful firmware. At last they're all cheap I guess.

P.S. I have several of the same basic Seagate drive you have (160GB, even
bought at CompUSA!) and would expect at least 2-3X better pgbench results
than you're seeing. I realized that I've never actually run that test
without first tweaking the postgresql.conf
(shared_buffers,checkpoint_segments) so that may be part of it. One of my
systems here has just one of those disk in it, next time I boot that up
I'll see what results I get with an untuned config.

--
* Greg Smith gsmith(at)gregsmith(dot)com http://www.gregsmith.com Baltimore, MD