[gmx-users]future hardware

[Andrey V. Golovin]

Interesting, will Intel use Xbox Xeon for servers and  workstations or 
it will be just a console version? AMD actually planing quad core 
opterons and more cores on chip 
roadmaphttp://www.ixbt.com/short/images/tech_roadmap.gif ..
so we need benchmarks =)
Andrey

> Hi,
>
> Fujitsu have already created a Gromacs-based MD server ("bioserver") 
> with embedded low-power CPUs which I think is better for that type of 
> application.
>
> We've been interested in Cell for a while, but there's a lot of 
> marketing hype around it. First, it's not 9 separate processors but 1 
> main processor and 8 somewhat independent vector units. The vector 
> units can't access memory directly though, but need to be fed with 
> programmed DMA from the head CPU. There is also no dynamic instruction 
> scheduling.
>
> From a Gromacs perspective I think the Xenon 3-way standard SMP 
> PowerPC CPU (but still no dynamic instruction scheduling) in the 
> Xbox360 is more interesting, especially since it seems to use standard 
> Altivec instructions that we've already implemented. Unfortunately we 
> can't test that until somebody has cracked the Xbox360 and put Linux 
> or something on it :-)
>
> Cheers,
>
> Erik
>
>
> On Dec 1, 2005, at 10:11 AM, Andrey V. Golovin wrote:
>
>> Hi all !
>> I just recently found that NEC gonna produce MD extension cards:
>> http://www.nec.co.jp/press/en/0511/2901-01.html , from other side new
>> cell processor
>> (http://www.mc.com/products/view/index.cfm?type=boards&id=64)
>> have less power then NEC device. Actually what are you guys think about
>> all this ? could it improve gromacs  performance? Cell system for  me
>> sounds quite ok because it just 8 powerpc in  chip.
>> what do you think?
>>
>> All the best !
>>
>> -------------------------------------------------------------------
>> Andrey V. Golovin
>> Ph.D,Professor assistant    tel: (095) 939-3149
>> Bioengineering and
>> Bioinformatics Department
>> Moscow State University     fax: (095) 939-3181
>> 119899 Moscow               E-mail: golovin at genebee.msu.su
>> Russia                      web: http://rnp-group.genebee.msu.su
>> -----------------------------------------------------------------
>>
>>
>>
>> _______________________________________________
>> gmx-users mailing list
>> gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please don't post (un)subscribe requests to the list. Use the www 
>> interface or send it to gmx-users-request at gromacs.org.
>>
>
> -----------------------------------------------------------
> Erik Lindahl  <lindahl at sbc.su.se>
> Assistant Professor, Stockholm Bioinformatics Center
> Stockholm University, SE 106 91 Stockholm
> Phone: +46 8 5537 8564     Fax: +46 8 5537 8214
>
>
>
>
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use thewww 
> interface or send it to gmx-users-request at gromacs.org.
>


-- 
-----------------------------------------------------------------
Andrey V. Golovin
Ph.D,Professor assistant    tel: (095) 939-3149
Bioengineering and
Bioinformatics Department
Moscow State University     fax: (095) 939-3181
119899 Moscow               E-mail: golovin at genebee.msu.su
Russia                      web: http://rnp-group.genebee.msu.su
----------------------------------------------------------------- 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051201/90ac486b/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tech_roadmap.gif
Type: image/gif
Size: 21651 bytes
Desc: not available
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051201/90ac486b/attachment.gif>