The ABINIT Users Mailing List ( CLOSED )

[mmikami@rc.m-kagaku.co.jp]

Hello,

I have not tried Opteron boxes, but I am curious whether the Absoft 
Fortran compiler will work well or not for such purpose : 
http://www.absoft.com/opteron.html
Kindly see also the links therein: i.e.,
http://www.absoft.com/productinfo_opteron.html#comparison
http://www.absoft.com/literature/mpich_guide.txt

Parenthetically, I note :
I have tried the Absoft Fortran compiler (v.8.0 and v.8.2) for MacOS X 
(v.10.2.8). So far, v.8.0 worked OK, but v.8.2 could not compile ABINIT 
v.4.2.3 successfully. I am contacting the Absoft support desk. Let us 
wait and see...

In passing,  I also tried IBM XL Fortran for MacOS X (v.10.2.8) as well:
http://www-306.ibm.com/software/awdtools/fortran/xlfortran/support/  or
http://www14.software.ibm.com/webapp/download/search.jsp?go=y&rs=xlf3
(this is beta-release, so it can be freely available so far after 
registration. This compiler may work on G3/G4 as well as G5; My 
PowerMac is actually G4, beyond the hardware requirement. But I am not 
sure it will work on MacOSX 10.3.x ("Panther"))
The Test_* results seems very similar with those with the Absoft 
ProFortran v.8.0, most cases passed successfully  except several 
cases...
I sent my makefile_macros for XL fortran for MacOSX, so it might be 
included in src_tests*.tar.gz in future ... (If I successfully 
compile/run the OpenMP version later, I will make a report...)

Regards,
Masayoshi

On 2003.12.26, at 08:10 Asia/Tokyo, Artem R. Oganov wrote:

> Dear all,
>
> We are now thinking about buying a PC Linux cluster based on Opteron
> 1.8GHz processors each with 8GB RAM  (probably with up to 64 processors
> in our cluster). The processors will be connected by a fast network
> (probably 1Gbit). Can anyone comment on such a machine? In particular, 
> I
> would like to know what can be potential problems, how suitable would
> the machine be for parallel calculations, how many processors can be
> usefully run for one job, the best compiler and libraries, etc. -
> whatever is of importance.
> I apologise for a pretty big question - but I am sure there are many
> people who are in a similar situation and would like to know the 
> answer.
>
> Finally, let me wish you all happy holidays and a successful 2004!
>
> With my best regards,
>
> Artem
> ************************
> Dr. Artem R. Oganov
> Lab. of Crystallography, Department of Materials
> ETH Zurich, CH-8092 Zurich, Switzerland
> Tel.: +41-1-632-3752; Fax: +41-1-632-1133
> E-mail: a.oganov@mat.ethz.ch
> Personal web-page: http://olivine.ethz.ch/~artem
> Group page:  http://www.crystal.mat.ethz.ch/research/ExtremeConditions
> ************************