From: Nicholas M Glykos (glykos_at_mbg.duth.gr)
Date: Wed Apr 14 2010 - 05:01:06 CDT
Just guessing, but: Is the trajectory written on a local disk, or on a
NFS-mounted volume ? If it is NFS, it could be that it has been mounted
with the async option, in which case issuing a 'sync' on both the server
and client side may solve your worries.
On Wed, 14 Apr 2010, francesco oteri wrote:
> Dear NAMD users,
> I'm running 2 simulations of 10ns on 2 different intel quad core PC with
> Ubuntu 9.10 4GB RAM each one, using binary cuda version (CVS 2010-03-26)
> with 2cpu on a "GeForce GTS 250" . The folder containing namd is hosted on a
> server and the 2 pc access to it through NFS.
> I'm experiencing a strange behaviour: though namd seems to run (how
> displayed by top command) after a random number of steps, no data are
> written to disk.
> I've started the simulations 10 days ago and both are still running (
> though should be finished this night) , but the first displayed the problem
> after 1day (around 900000 steps) while the second displayed the problem this
> morning after 4858000 steps.
> I observed that Xorg use 100% of a CPU, but killing it doesn't solve the
> problem; the first simulations shows the problem more often that the second.
-- Dr Nicholas M. Glykos, Department of Molecular Biology and Genetics, Democritus University of Thrace, University Campus, Dragana, 68100 Alexandroupolis, Greece, Tel/Fax (office) +302551030620, Ext.77620, Tel (lab) +302551030615, http://utopia.duth.gr/~glykos/
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:55:41 CST