Dear Andreas,
Thanks for your suggestions. The machine will be intented to use
sander.MPI for TI calculations. I will use 2 CPUs for each job. We are
thinking either to buy 1 machine which has two Quad Core in it, or 2
machines which has two dual core in it each. Maintaining one machine will
be easier for me, so I prefer the two Quad Core machine; but if the
efficiency of running sander.MPI is better in two Dual Core machines, we
will go with the second choice.
I have never run any jobs in a machine which had two Quad Core in it, so I
am wondering about it; whether it is as efficienct (or maybe better!?)
as the two Dual Core machines. Thanks in advance.
Best,
On Fri, 3 Aug 2007, Andreas Svrcek-Seiler wrote:
> Hi,
> > We are going to buy the following DELL system to our group to run AMBER
> > 9: Dell Precision Workstation 490n MiniTower, running in Red
> > Hat Enterprise Linux WS v4. It has two Quad Core Intel Xeon Processor
> > E5335 (2 GHz, 2X4MB L2, 1333). It is going to have 8 GB of RAM in it.
> > I was wondering if any of you had any experience on running AMBER in this
> > kind or similar system; particularly the efficiency of running sander.MPI.
> > Is 8 GB of RAM enough, or can we get better speed by increasing the RAM to
> > 16 GB?
> No. If your system is so big that you need more than 8 GB of RAM, your job
> would run unbearably slowly anyway.
> Some time ago I tried the "one million atom" benchmark (see
> http://www.ks.uiuc.edu/Research/namd/utilities/) from the NAMD
> developers on several nodes (using NAMD of course).
> Each node had 4 GB of RAM and none started swapping ( =running out of RAM
> and using the harddisk instead).
>
> ...just my 0.02$ and others might have a different opinion.
>
> good luck
> Andreas
>
> P.S.: We're going to get just such a machine and I'm really looking
> forward to benchmarking such an expensive toy :-)
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
>
--
Ilyas Yildirim
---------------------------------------------------------------
- Department of Chemistry - -
- University of Rochester - -
- Hutchison Hall, # B10 - -
- Rochester, NY 14627-0216 - Ph.:(585) 275 67 66 (Office) -
- http://www.pas.rochester.edu/~yildirim/ -
---------------------------------------------------------------
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Sun Aug 05 2007 - 06:07:54 PDT
