"Chad C. Sines" wrote:
>> Our lab is strongly considering purchasing a cluster of ~12 PC with dual PII 450 MHz processors and 256 MB RAM each to configure in a Beowulf cluster running Linux. Does anyone have experience running this type of system? Reading over past postings, I have come across people mentioning "possible" configurations, but I have never read where anyone has actually successfully ran CNS over a cluster. I know that some software, such as Shelx, is made to run on only one CPU. What about CNS? I would be very interested in hearing either positive or negative comments from people who have experience running CNS on a Beowulf cluster.
>> Chad
Hi,
I'm fairly familiar with the CNS codebase--I wrote a
large chunk of it. I'd be rather surprised if you could
get it working efficiently on a rather large-scale
parallel machine. The internals just aren't designed
for that.
However, if you just want to be able to run several
independent structures at once (as is commonly done
in NMR structure determination), that should be
possible, and even easy with some of the CNS
extensions to the input language.
What sort of calculations are you intending to do
on that Beowulf monster? It seems to me that most
people (ie., crystallographers) find xplor/CNS fast
enough on modern single processor machines.
--
_____________
| ___/_
| |/ /
-- /\ // /--
Dr. John Kuszewski || || / /||
johnk at speck.niddk.nih.gov || || / / ||
|| ||/ / ||
|| |/ /| ||
|| / /|| ||
My parents went to Zaire \/ / / || \/
and all I got |/__/| |
was this lousy retrovirus. /_________|
