NAMD
NAMD is an award-winning parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. Based on Charm++ parallel objects, NAMD scales to hundreds of cores for typical simulations and beyond 500,000 cores for the largest simulations. NAMD uses the popular molecular graphics program VMD for simulation setup and trajectory analysis, but is also file-compatible with AMBER, CHARMM, and X-PLOR.
Useful Links
Using NAMD on ARCHER2
NAMD is freely available to all ARCHER2 users.
Running parallel NAMD jobs
Note
The simplest approach, and the one that is likely to give the best performance, is to use the pure MPI version of NAMD (i.e. nosmp mode) which involves loading a non-default NAMD module.
The following script will run a pure MPI NAMD MD job using 4 nodes (i.e. 128x4 = 512 CPU-cores).
#!/bin/bash
# Request four nodes to run a job of 512 MPI tasks with 128 MPI
# tasks per node, here for maximum time 20 minutes.
#SBATCH --job-name=namd-nosmp
#SBATCH --nodes=4
#SBATCH --tasks-per-node=128
#SBATCH --cpus-per-task=1
#SBATCH --time=00:20:00
# Replace [budget code] below with your project code (e.g. t01)
#SBATCH --account=[budget code]
#SBATCH --partition=standard
#SBATCH --qos=standard
module load namd/2.14-nosmp
srun --distribution=block:block --hint=nomultithread namd2 input.namd
#!/bin/bash
# Request four nodes to run a job of 512 MPI tasks with 128 MPI
# tasks per node, here for maximum time 20 minutes.
#SBATCH --job-name=namd-nosmp
#SBATCH --nodes=4
#SBATCH --tasks-per-node=128
#SBATCH --cpus-per-task=1
#SBATCH --time=00:20:00
# Replace [budget code] below with your project code (e.g. t01)
#SBATCH --account=[budget code]
#SBATCH --partition=standard
#SBATCH --qos=standard
# Setup the job environment (this module needs to be loaded before any other modules)
module load epcc-job-env
module load namd/2.14-nosmp-gcc10
srun --distribution=block:block --hint=nomultithread namd2 input.namd
If your jobs runs out of memory, then you can can run the smp version of NAMD which uses less memory. This involves launching a combination of MPI processes for communication and worker threads which perform computation.
Note
The optimal choice for the numbers of MPI processes and worker threads depends on the dataset and the number of compute nodes. Before running large production jobs, it is worth experimenting with these parameters to find the optimal configuration for your particular simulation.
The following script will run an SMP NAMD MD job using 4 nodes (i.e.
128x4 = 512 CPU-cores) with 32 MPI communication processes per node
and 96 worker threads. The number of workers per communication process
is specified by the +ppn argument to NAMD, which is set here to
cpus-per-task - 1, i.e +ppn 3, to leave a CPU-core free for the
associated MPI process.
#!/bin/bash
#SBATCH --job-name=namd-smp
#SBATCH --tasks-per-node=32
#SBATCH --cpus-per-task=4
#SBATCH --nodes=4
#SBATCH --time=00:20:00
# Replace [budget code] below with your project code (e.g. t01)
#SBATCH --account=[budget code]
#SBATCH --partition=standard
#SBATCH --qos=standard
# Load the relevant modules
module load namd
# Set procs per node (PPN) & OMP_NUM_THREADS
export PPN=$(($SLURM_CPUS_PER_TASK-1))
export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
export OMP_PLACES=cores
# Record PPN in the output file
echo "Number of worker threads PPN = $PPN"
# Run NAMD
srun --distribution=block:block --hint=nomultithread namd2 +setcpuaffinity +ppn $PPN input.namd
#!/bin/bash
#SBATCH --job-name=namd-smp
#SBATCH --tasks-per-node=32
#SBATCH --cpus-per-task=4
#SBATCH --nodes=4
#SBATCH --time=00:20:00
# Replace [budget code] below with your project code (e.g. t01)
#SBATCH --account=[budget code]
#SBATCH --partition=standard
#SBATCH --qos=standard
# Load the relevant modules
module load epcc-job-env
module load namd
# Set procs per node (PPN) & OMP_NUM_THREADS
export PPN=$(($SLURM_CPUS_PER_TASK-1))
export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
export OMP_PLACES=cores
# Record PPN in the output file
echo "Number of worker threads PPN = $PPN"
# Run NAMD
srun --distribution=block:block --hint=nomultithread namd2 +setcpuaffinity +ppn $PPN input.namd
This is likely to be a reasonable first choice but you should
experiment with the values of tasks-per-node and cpus-per-task. To
use all the CPU-cores in a node, the product of these two values
should be 128.
If you cannot run the pure MPI version of NAMD, then the optimal
values of (tasks-per-node, cpus-per-task) are likely to be either
(32,4), (16,8) or (8,16).
Compiling NAMD
The latest instructions for building NAMD on ARCHER2 may be found in the GitHub repository of build instructions.