Re: pg 8.1.3, AIX, huge box, painfully slow.

On Fri, 07 Apr 2006 13:54:21 -0500
Scott Marlowe <smarlowe(at)g2switchworks(dot)com> wrote:

> Are the same queries getting the same basic execution plan on both
> boxes? Turn on logging for slow queries, and explain analyze them on
> both machines to see if they are.

See reply to Tom Lane :)

> I'd put the old 4 way Xeon back in production and do some serious
> testing of this pSeries machine. IBM should be willing to help you, I
> hope.

They probably would if this had been bought new - as it is, we have
rented the machine for a month from a 2nd-user dealer to see if it's
capable of taking the load. I'm now glad we did this.

> My guess is that this is an OS issue. Maybe there are AIX tweaks that
> will get it up to the same or higher level of performance as your four
> way xeon. Maybe there aren't.

The pSeries isn't much older than our Xeon machine, and I expected the
performance level to be exemplary out of the box.. we've enabled the
64-bit kernel+userspace, and compiled pg for 64-bitness with the gcc
flags as reccommended by Senica Cunningham on this very list..

> Myself, I'd throw a spare drive in for the OS, put some flavor of
> linux on it

Terrifying given I know nothing about the pSeries boot system, but at
this stage I'm game for nearly anything.

> http://www.asaservers.com/system_dept.asp?dept_id=SD-002

Multi-Opteron was the other thing we considered but decided to give
'Big Iron' UNIX a whirl...

Cheers,
Gavin.