namd

This is an old revision of the document!


NAMD, recipient of a 2002 Gordon Bell Award, is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD is file-compatible with AMBER, CHARMM, and X-PLOR and is distributed free of charge with source code. Documentation and source code on the web page of “Theoretical and computational biophysics group

There are few possibilities to start NAMD simulation on the MOGON cluster.

<PATH TO NAMD INSTALLATION>/namd2 <input configuration file>

This starts one threaded simulation direkt on the login host. Input configuration file describes simulation. Complete description of the input file format is available in the NAMD Documentation.

Threaded jobs

Using threaded start allows to achieve performance increasing:

bsub -R “affinity[core(<number of cores>)]” <PATH TO NAMD INSTALLATION>/namd2 +p<number of threads> <input configuration file>

where <number of threads> is normally equal to <number of cores>. For example

bsub -R “affinity[core(16)]” <PATH TO NAMD INSTALLATION>/namd2 +p 16 <input configuration file>

starts NAMD simulation with 16 threads using 16 cores of a node.

Mogon cluster contains different nodes with up to 64 cores.

On machines with more than 32 cores it may be necessary to add a communication thread and run on one fewer core than the machine has. For example on a 64-core machine this would be run as

bsub -R “affinity[core(64)]” <PATH TO NAMD INSTALLATION>/namd2 +p 63 +commthread <input configuration file>

Memory Usage

NAMD has traditionally used less than 100MB of memory even for systems of 100,000 atoms. With the reintroduction of pairlists in NAMD 2.5, however, memory usage for a 100,000 atom system with a 12A cutoff can approach 300MB, and will grow with the cube of the cutoff. This extra memory is distributed across processors during a parallel run, but a single workstation may run out of physical memory with a large system. Appropriate method for memory reservation is using of application profile, for example: start 1 NAMD process with 64 calculation threads in 15G memory pool

bsub -app Reserve15G -R “affinity[core(64)]” <PATH TO NAMD INSTALLATION>/namd2 +p 64 <input configuration file>

or with reservation 5000000 Kb of memory for this 1 process

bsub -M 5000000 -R “affinity[core(64)]” <PATH TO NAMD INSTALLATION>/namd2 +p 64 <input configuration file>

non threaded MPI jobs

bsub -n <number of processes> -M 1000000 -R “affinity[core(1)]” mpirun <PATH TO NAMD INSTALLATION>/namd2 <input configuration file>

Using of affinity and core mapping

There are two possibility to set process to core mapping.

bsub -n <number of processes> -M 1000000 -R “affinity[core(1)*2]” mpirun <PATH TO NAMD INSTALLATION>/namd2 <input configuration file>

…under construction …

  • namd.1403009997.txt.gz
  • Last modified: 2014/06/17 14:59
  • by noskov