Panel | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||
Amber is the collective name for a suite of programs that allow users to carry out molecular dynamics simulations, particularly on biomolecules. The Amber home page can be found at ambermd.org |
Panel | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||
Anchor | ||||
---|---|---|---|---|
|
To use this version of Amber set the environment
Code Block | ||
---|---|---|
| ||
$ module load amber/20 |
To understand more about modules see our software applications guide.
Amber 20 supports both serial and parallel versions of the programs. Parallel versions have .MPI suffix, for example, pmemd.MPI, sander.MPI, sandel.LES.MPI. Serial versions of these programs are pmemd, sander, sander.OMP (to run sunder in parallel inside one node) and sander.LES.
Amber pmemd program can also use GPUs. Corresponding executables are pmemd.cuda, pmemd.cuda_DFDP, pmemd.cuda_SPFP, pmemd.cuda_DFDP.MPI, pmemd.cuda_SPFP.MPI. Indecies DFDP and SFDP stands for double and single precision floating points revisions.
The parallel codes must be executed via an mpirun command.
The serial analysis programs and nmode simulation program should only be submitted with a request for 1 cpu.
Here is a sample job script bench.dhfr for sander from the directory $AMBERHOME/benchmarks/dhfr using project a99 with 10 minutes walltime and 800 MiB memory over 4 CPUs.
Code Block | ||
---|---|---|
| ||
#!/bin/bash #PBS -l walltime=10:00 #PBS -l mem=800mb #PBS -l ncpus=4 #PBS -l software=amber #PBS -l wd # Load module, always specify version number. module load amber/20 # Must include `#PBS -l storage=scratch/ab12+gdata/yz98` if the job # needs access to `/scratch/ab12/` and `/g/data/yz98/` cat <<eof >mdin equilbrate apo-dhfr &cntrl ntb=1, ntp=0, ntpr=100, ntc=2, ntf=2, tol=0.000001, nstlim=100, ntt=0, dt=0.001, ntx=7, irest=1, vlimit=10.0, / eof mpirun sander.MPI -O -i mdin -c md12.x -o bench.dhfr.out |
Submit to queueing system:
Code Block | ||
---|---|---|
| ||
$ qsub ./bench.dhfr |
Using the GPU queue Anchor Using the GPU queue Using the GPU queue
Code Block | ||
---|---|---|
| ||
#!/bin/bash #PBS -q gpuvolta #PBS -l walltime=10:00 #PBS -l mem=800mb #PBS -l ncpus=12 #PBS -l ngpus=1 #PBS -l software=amber #PBS -l wd # Load modules, always specify version number. module load cuda/10.1 module load amber/20 # Must include `#PBS -l storage=scratch/ab12+gdata/yz98` if the job # needs access to `/scratch/ab12/` and `/g/data/yz98/` mpirun -np $PBS_NGPUS pmemd.cuda.MPI -O -i mdin -o mdout -inf mdinfo -x mdcrd -r restrt |
Anchor | ||||
---|---|---|---|---|
|
The AmberTools suite is free of charge, and its components are mostly released under the GNU General Public licence (GPL). A few components are included that are in the public domain or which have other, open-source, licences. The sander program now has the LGPL licence.
Access to this version of amber is available to users from non-profit organizations (including CSIRO). Commercial use of amber will require prior discussion with the National Facility and have a separate licence arrangement with the Regents of the University of California.
Please note that the software should be referenced in publications.
Please request to join software group on https://my.nci.org.au if you wish to request access.