Re: Hardware for NAMD / VMD machine

From: Axel Kohlmeyer (akohlmey_at_cmm.chem.upenn.edu)
Date: Fri Jul 03 2009 - 11:56:08 CDT

On Fri, 2009-07-03 at 00:26 -0600, Olve Peersen wrote:

olve,

> We're looking into buying a new Linux box for structural biology
> refinement and some molecular dynamics work (i.e. equilibration runs
> and making sure input files are right before handing jobs off to
> bigger clusters) and I wanting to get some input on where it is worth
> spending money on the system from a NAMD/VMD perspective. Our
> experimental systems are proteins of 150-500 residues in size, and we
> already have some experience running NAMD/VMD on an older Pentium 4D
> system with a Quadro FX1300 and hardware stereo.

> At the moment I am leaning toward a dual Xeon 55XX (quad-core) system
> with 12 GB of RAM - this is the Xeon version of the quad-core Core i7
> Nehalem processor. My choice will probably end up being the midrange
> 5520 or 5530 processors (2.26 vs 2.40 GHz) unless there is a really
> good computational reason to go to the faster X series, where I
> suspect I'll be mostly paying a "bleeding-edge premium".

that seems to be a very reasonable choice. for NAMD itself, the quantity
of memory doesn't matter so much, but VMD will certainly benefit from
large memory (i'd probably go for 24Gig).

> For NAMD it seems that more and faster processors, more RAM, and fast
> disks are the key elements. I'm not up for doing full cluster
> architecture, so any inter-machine connections would be gigabit for

one more option to consider would be to make sure you have the best
and largest possible power supply and then add a couple of nvidia
GTX 285 cards or even some tesla C1060. both NAMD and VMD can be
compiled to support GPU acceleration. in my last test on a nehalem
type machine with two C1060 GPUs, i get about the same NAMD performance
with two GPUs than what i get using all 8 cores, and that was on a
machine with only 8x PCIe slots. on a proper desktop with 16x PCIe
you should get even better performance.

> the time being. I also suspect I'd be better off compiling with the
> Intel compilers and certain Xeon specific flags as some processor
> specific tweaking is apparently worthwhile for the new Nehalem
> architecture, but that is something I have not seriously looked into
> yet.

the main advantage of nehalem over the previous architecture is
its 3x larger memory bandwidth. NAMD doesn't benefit very much
from it, since the data sets per core are generally comparatively
small.

> The question I've had a harder time nailing down is how dependent VMD
> performance is on the graphics card versus the core system itself and
> how powerful a 3D graphics card I should invest in. In our experience
> thus far, rendering of our systems seems fine with our old Quadro
> FX1300 and I was thinking of going with a 768 MB Quadro FX1800 for

i would stay away from quadro cards if you can. their price tag is
very high for the performance they deliver. in comparison to that
the GT200 series "consumer"-type cards are usually better performing
at a _much_ lower price.

> this new system, but I could spend another $400 and get a 1GB FX3800,
> or get really carried away with ~$1700 for a FX4600/4800 card. Is
> there a compelling performance reason for doing so?

a GTX285 for under $400 or less has been the fastest GPU that we've
been able to get our hands on so far (note the GTX295 are basically
two GTX260 cards with slightly lowered clock glued into one case).

the nehalem machines seem to be a good match to adding GPUs for
computing, as the GPUs benefit from the large memory bandwidth.

cheers,
   axel.

> Thanks in advance for any input,
>
> Olve
>
> -------------------------------------------------------
> Olve Peersen
> Associate Professor
> Dept. of Biochemistry & Molecular Biology
> 1870 Campus Delivery
> Colorado State University
> Ft. Collins, CO 80523-1870
> -------------------------------------------------------
> 970.491-0433 Office
> 970.491-0271 Lab
> 970.491-0494 Fax
> -------------------------------------------------------
>

-- 
=======================================================================
Axel Kohlmeyer   akohlmey_at_cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:59 CST