[gmx-users] linux cluster performance
spoel at xray.bmc.uu.se
Mon May 2 18:01:49 CEST 2005
On Mon, 2005-05-02 at 20:55 +0530, Alok wrote:
> Hi Gromaxians,
> sorry for unrelated question.
> My lab is going to purchase a Linux Cluster and we have a couple of
> queries regarding the cluser.
> 1) We have heard about "XEON Based cluster".Can anyone shed some light
> into this form of the cluster and how it differs from a more
> conventional PCB(Personel Computer Based) Linux Cluster. Which of the
> two would be a better option under normal circumstances???
> 2) The other query is regarding the scaling i.e.speed wise performance
> of the cluster. We heard that after a certain no of nodes increment
> ( say upto 8-10), the performance (speed) of the cluster is not
> greatly effected or the speed performance of the cluster reaches a
> plateau even if we increase the number of nodes.
> If so is the case, would it be sensible to go for say a two
> different 8 node cluster instead of a say one 16 node cluster.
> I will be thankfull if some one suggest somthing regarding the above
> mentioned issues.
I have just gone through negotiations for a new cluster and will buy a
cluster consisting of mainly dual-core opteron 2-way machines. This
gives you four processors per box which you can use for either efficient
4-processor jobs or multiple single- or two-processor jobs.
You can get these from big companies or your favorite white box dealer.
> Alok Jain
> gmx-users mailing list
> gmx-users at gromacs.org
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596, 75124 Uppsala, Sweden
phone: 46 18 471 4205 fax: 46 18 511 755
spoel at xray.bmc.uu.se spoel at gromacs.org http://xray.bmc.uu.se/~spoel
More information about the gromacs.org_gmx-users