[gmx-users] about parallel work

> Hello, every body. I have a question about parallel running of mdrun_mpi. I > doubt that sometimes the parallel running of mdrun_mpi can not utilize the > domain decomposition. > This is the case when I use the batch work in the computer cluster which > install the MPICH2. In the case, I use comm

Re: [gmx-users] about parallel work

Am 21.04.2009 um 17:04 schrieb sheerychen: ... However, in my person computer which install the MPICH (not MPICH 2), I use the commond like this ''mpirun -np 3 /usr/bin/mdrun_mpi - deffnm *** -v''. It would show the domain decomposition and the speed is quicked than 8 CPUS. What is the pro

Re: [gmx-users] about parallel work

yes, both versions are compiled as mpi version. However the start mpi messages are different. For MPICH, it would show that 1D domain decomposition like 3*1*1, and only 1 file would be produced. However for MPICH2, no such information appears and it would produce many files as (8 nodes): complex_em

Re: [gmx-users] about parallel work

On Apr 21, 2009, at 5:04 PM, sheerychen wrote: Hello, every body. I have a question about parallel running of mdrun_mpi. I doubt that sometimes the parallel running of mdrun_mpi can not utilize the domain decomposition. This is the case when I use the batch work in the computer cluster

Re: [gmx-users] about parallel work

Hi, On Apr 21, 2009, at 5:53 PM, sheerychen wrote: yes, both versions are compiled as mpi version. However the start mpi messages are different. For MPICH, it would show that 1D domain decomposition like 3*1*1, and only 1 file would be produced. However for MPICH2, no such information appe

[gmx-users] Ewald summation: n=0

Gromacs Users: While porting field parameters to Gromacs, I ran across unexpected electrostatic energies returned by the ewald summation method. To simplify the problem, I reduced the system to two partially charged particles. For the simplest case of only one box, n = 0, I expected to see

[gmx-users] Continuous trajectory processing

Hello, Thanks for reading this. I'm doing many-nanosecond simulations (GMX 3.3.1) with a 2 fs time step and I would like to analyze every single frame (or at minimum every tenth frame) of the trajectory for the presence of a hydrogen bond to one particular atom in the simulation. The problem with

[gmx-users] identical energies in a rerun calculation

Hello, I am trying to do a -rerun simulation in Gromacs 4.0.4 to calculate the interaction energy between 2 residues using the trajectory I already had from the previous simulation: /gpfs/apps/GROMACS/4.0.4/bin/mdrun -v -deffnm E_interaccion_Asp -dlb auto -rerun equilibrado3_19.xtc

Re: [gmx-users] Continuous trajectory processing

you have the choice of using the -dt XX option of g_hbond which will tell to analyze only every XX structure. I am not sure this is present in g_hbond. the other way is to use trjconv with the dt option again an write the frames every XX steps. On Apr 21, 2009, at 9:26 PM, Aaron Fafarman wro

[gmx-users] about the bond connection between different groups

Hi all users, I wonder whether it is allowed to define the bond connection between different group in gromacs. Suppose atom A is in group1 and there is another atom B in group2 .Then, I want to define bond between atom A and B. I am not sure whether this is available in gromcas. Thank you for

Re: [gmx-users] about the bond connection between different groups

He, Yang wrote: Hi all users, I wonder whether it is allowed to define the bond connection between different group in gromacs. Suppose atom A is in group1 and there is another atom B in group2 .Then, I want to define bond between atom A and B. I am not sure whether this is available in grom

RE: [gmx-users] about the bond connection between different groups

Hi Justin, It seems that I can not fix the fragmentation fault when position restraints is added and hence, I just consider using the freezegroup method to make some atoms in a whole group fixed .While the freezegroup seems to just be useful for the whole group not applicable to some certain a

Re: [gmx-users] about the bond connection between different groups

He, Yang wrote: Hi Justin, It seems that I can not fix the fragmentation fault when position restraints is added and hence, I just consider using the freezegroup method to make some If all trials of PR fail, even test cases with known systems, then you need to consider that your installatio

Re: [gmx-users] Continuous trajectory processing

Aaron Fafarman wrote: Hello, Thanks for reading this. I'm doing many-nanosecond simulations (GMX 3.3.1) with a 2 fs time step and I would like to analyze every single frame (or at minimum every tenth frame) of the trajectory for the presence of a hydrogen bond to one particular atom in the simul

Re: [gmx-users] about the bond connection between different groups

Hi He Yang, Justin, >> You have said bonds between distinct molecules require a merged topology. >> Is >> there any introduction in the manual or Do you have any example about the >> merged topology? >> > > A merged topology contains multiple moleculetype definitions in one > topol.top. Discussion