[gmx-users] segmentation fault in mdrun when using PME

[Diane Fournier]

Hi, I'm new to Gromacs and I'm trying to run a enzyme-ligand complex molecular dynamics simulation. I have tried doing John Kerrigan's Drug-Enzyme tutorial and mdrun crashes with segmentation fault and core dump at the steepest descents minimization step. However, mdrun works fine when using cutoff instead of PME. 
 
I'm working with Gromacs v. 3.3.1 on a SGI altix 3700 with 32 Intel Itanium 2 processors (but I'm currently using a single node, so it's not a MPI problem) under Red Hat Enterprise Linux AS release 3 with Intel Math Kernel Librarary (MKL) v. 8.0.1 as FFT library (which is optimized for Itanium 2).
 
the em.mdp file looks like:
 
title               =  drg_trp
cpp                 =  /usr/bin/cpp
define              =  -DFLEX_SPC
constraints         =  none
integrator          =  steep
dt                  =  0.002    ; ps !
nsteps              =  500
nstlist             =  10
ns_type             =  grid
rlist               =  0.9
coulombtype             =  PME
rcoulomb            =  0.9
rvdw                =  0.9
fourierspacing          =  0.12
fourier_nx              =  0
fourier_ny              =  0
fourier_nz              =  0
pme_order               =  4
ewald_rtol              =  1e-5
optimize_fft            =  yes
;
;       Energy minimizing stuff
;
emtol               =  1000.0
emstep              =  0.01

Is it possible this could be related to insufficient memory allocation ? How demanding is this PME calculation ?
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20060428/15bdec3c/attachment.html>