Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 28 Jan 2006 18:50:05 -0200
From:      JoaoBR <joao@matik.com.br>
To:        "O. Hartmann" <ohartman@mail.uni-mainz.de>
Cc:        kono@kth.se, Daniel Rock <freebsd@deadcafe.de>, mumag@nist.gov, freebsd-amd64@freebsd.org
Subject:   Re: dual vs single core opteron 100's
Message-ID:  <200601281850.07422.joao@matik.com.br>
In-Reply-To: <43DB7457.6050605@mail.uni-mainz.de>
References:  <200601270957.44032.kono@kth.se> <200601280912.52907.joao@matik.com.br> <43DB7457.6050605@mail.uni-mainz.de>

next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday 28 January 2006 11:40, O. Hartmann wrote:
> >
> > If I am not wrong there is actual only the X2-4800 and 4400 with 2x1MB
> > cache, all others are having max 512K
>
> We were talking about comparable Athlon64/Opterons for socket 939.
>
> > The Opterons are having 2MB of cache, not 2x1
>
> No, that's wrong. Dual-core 1XX Opterons do have two times 1MB 2nd level
> cache for each core, not a unified 2MB cache!
>

right, well, I didn't read with attention ... sincerly I was not even aware=
 of=20
existing socket 939 opterons so all my opteron opinions were not related wi=
th=20
this ones

I just reread the processor comparism query on amd's site where the dual=20
opterons are returned with 2MB cache=20
http://www.amdcompare.com/us-en/opteron/Default.aspx#grid

but on the data sheet they are as 1MB per core, so 2x1Mb
http://www.amd.com/us-en/Processors/ProductInformation/0,,30_118_8796_9240,=
00.html

I guess you are right here, make sense and I do not care so much either ;)

anyway, amd should be able to give exact info on all papers the publish

> All right, talking about scientific calculations, I would prefer a real
> dual chip plus dual core Opteron 2xx system due to the fact, memory
> bandwith is much more efficient when using NUMA architecture. But in my
> opinion, comparin dual chip and dual core Opterons (and their Athlon64
> counterparts) is comparing cherries and pineapples ...
>

sure, absolutly, but often on smaller or low bandwidth server the differenc=
e=20
is not even appearing, for the client I mean, I have no idea how this is on=
=20
PCs or Workstations

but money is a strong point when you get the same quantity of "cherries" at=
=20
the end, so the comparism is it worth to talk about and find out the limits


Jo=E3o







A mensagem foi scaneada pelo sistema de e-mail e pode ser considerada segura.
Service fornecido pelo Datacenter Matik  https://datacenter.matik.com.br



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200601281850.07422.joao>