Steinar H. Gunderson wrote:
> On Sun, Nov 11, 2007 at 08:27:02PM +0100, Dimitri wrote:
>> As well to demonstrate OS parallelism it's not so impressive to see
>> 4CPU server results rather 8CPU or 32threaded Niagara... Don't know
>> why they did not present similar performance graphs for these
>> platform, strange no?...
> I guess it's because their Niagara support is still very raw, and besides,
> it's not a very common platform.
> /* Steinar */
Not sure how much coding would need to be done for Niagra chips but I
would think that it is more likely a problem of getting the funds so
they can have one to work on.
Get Sheeky @ http://Sheeky.Biz
In response to
pgsql-performance by date
|Next:||From: Dave Cramer||Date: 2007-11-12 14:51:30|
|Subject: difference between a unique constraint and a unique index ???|
|Previous:||From: Cédric Villemain||Date: 2007-11-12 10:17:57|
|Subject: Re: work_mem and shared_buffers|