| The CSF2 has been replaced by the CSF3 - please use that system! This documentation may be out of date. Please read the CSF3 documentation instead. To display this old CSF2 page click here. |
NAMD
Overview
Parallel molecular dynamics code (home page).
Two versions are available on CSF:
- Version 2.7 is compiled from source on the CSF (very limited access).
- Version 2.9 is a binary install.
Restrictions on use
Due to license restrictions users must request access to the appropriate CSF Unix group, namd or namdbin. Most users can only be granted access to the binary version. Please email its-ri-team@manchester.ac.uk for access.
All users should read and follow the license, a copy of which can be found $NAMD_HOME/license.txt, but only once you have access to the module. The following points are of particular note:
- NAMD is owned by and copyrighted to the University of Illinois – this is detailed throughout the license agreement. Of most note in this respect are clauses 2., 4., and 5.
- NAMD is for academic, research and internal business purposes only, e.g. not for commercial use. A definition of commercial use is given in clause 7 of the license.
- Citation of the NAMD software must appear in any published work. See clause 6 and the NAMD website for the required text.
Set up procedure
The NAMD 2.9 (binary install) modulefile is loaded using:
module load apps/binapps/namd/2.9
The NAMD 2.7 (compiled on CSF) modulefile has a number of prerequisites. Hence you should load the following, for use with non-InfiniBand (GigE) connected nodes:
module load compilers/intel/fortran/11.1.064 module load libs/intel/mkl/10.2u3 module load mpi/intel-11.1/openmpi/1.4.3 module load apps/intel-11.1/namd/2.7
In all verisons executable is named: namd2
Requirements
- the MPI versions were compiled against the Gigabit Ethernet and are not compatible with the Infiniband implementation of MPI
- Compilation was done with Intel 11.1 compilers
Example
Load the appropriate modulefiles as listed above. Then create a jobscript in the directory containing your input files, for example:
#!/bin/bash #$ -S /bin/bash #$ -cwd #$ -V #$ -pe smp.pe 8 # Example: 8 cores in this parallel environment (max 12) # NSLOTS is automatically set to the number of cores specified on the PE line mpirun -n $NSLOTS namd2
Submit the job using
qsub jobscript
where jobscript is the name of your jobscript above.
Further info
- BBSRC benchmarking on HECToR etc: http://www.bbsrc.ac.uk/funding/facilities/facilities.aspx?#highperformancecomputing