Bureaucrats, cc_docs_admin, cc_staff, rsnt_translations
2,837
edits
NAMD: Difference between revisions
no edit summary
(Marked this version for translation) |
No edit summary |
||
| Line 49: | Line 49: | ||
<translate> | <translate> | ||
== Verbs jobs == <!--T:16--> | == Parallel CPU jobs == | ||
=== MPI jobs === <!--T:18--> | |||
'''NOTE''': MPI should not be used. Instead use OFI on Cedar and UCX on other clusters. | |||
=== Verbs jobs === <!--T:16--> | |||
<!--T:51--> | <!--T:51--> | ||
| Line 86: | Line 91: | ||
$CHARMRUN ++p $P ++nodelist $NODEFILE $NAMD2 +idlepoll apoa1.namd | $CHARMRUN ++p $P ++nodelist $NODEFILE $NAMD2 +idlepoll apoa1.namd | ||
}} | }} | ||
<translate> | <translate> | ||
== UCX jobs == <!--T:42--> | === UCX jobs === <!--T:42--> | ||
This example uses 80 processes in total on 2 nodes, each node running 40 processes, thus fully utilizing its 80 cores. This script assumes full nodes are used, thus <code>ntasks-per-node</code> should be 40 (on Béluga). For best performance, NAMD jobs should use full nodes. | This example uses 80 processes in total on 2 nodes, each node running 40 processes, thus fully utilizing its 80 cores. This script assumes full nodes are used, thus <code>ntasks-per-node</code> should be 40 (on Béluga). For best performance, NAMD jobs should use full nodes. | ||
| Line 115: | Line 119: | ||
<translate> | <translate> | ||
== UCX GPU jobs == <!--T:44--> | === OFI jobs === <!--T:53--> | ||
<!--T:54--> | |||
'''NOTE''': OFI versions will run '''ONLY''' on Cedar because of its different interconnect. | |||
</translate> | |||
{{File | |||
|name=ucx_namd_job.sh | |||
|lang="sh" | |||
|contents= | |||
#!/bin/bash | |||
#SBATCH --account=def-specifyaccount | |||
#SBATCH --ntasks 64 # number of tasks | |||
#SBATCH --nodes=2 | |||
#SBATCH --ntasks-per-node=32 | |||
#SBATCH -t 0:05:00 # time (D-HH:MM) | |||
#SBATCH --mem=0 # memory per node, 0 means all memory | |||
#SBATCH -o slurm.%N.%j.out # STDOUT | |||
<!--T:55--> | |||
module load StdEnv/2020 namd-ofi/2.14 | |||
srun --mpi=pmi2 namd2 stmv.namd | |||
}} | |||
<translate> | |||
== Single GPU jobs == <!--T:19--> | |||
This example uses 8 CPU cores and 1 GPU on a single node. | |||
</translate> | |||
{{File | |||
|name=multicore_gpu_namd_job.sh | |||
|lang="sh" | |||
|contents= | |||
#!/bin/bash | |||
# | |||
#SBATCH --cpus-per-task=8 | |||
#SBATCH --mem 2048 | |||
#SBATCH -o slurm.%N.%j.out # STDOUT | |||
#SBATCH -t 0:05:00 # time (D-HH:MM) | |||
#SBATCH --gres=gpu:1 | |||
#SBATCH --account=def-specifyaccount | |||
module load StdEnv/2020 | |||
module load cuda/11.0 | |||
module load namd-multicore/2.14 | |||
namd2 +p$SLURM_CPUS_PER_TASK +idlepoll apoa1.namd | |||
}} | |||
<translate> | |||
== Parallel GPU jobs == | |||
=== UCX GPU jobs === <!--T:44--> | |||
This example is for Béluga and it assumes that full nodes are used, which gives best performance for NAMD jobs. It uses 8 processes in total on 2 nodes, each process(task) using 10 threads and 1 GPU. This fully utilizes Béluga GPU nodes which have 40 cores and 4 GPUs per node. Note that 1 core per task has to be reserved for a communications thread, so NAMD will report that only 72 cores are being used but this is normal. | This example is for Béluga and it assumes that full nodes are used, which gives best performance for NAMD jobs. It uses 8 processes in total on 2 nodes, each process(task) using 10 threads and 1 GPU. This fully utilizes Béluga GPU nodes which have 40 cores and 4 GPUs per node. Note that 1 core per task has to be reserved for a communications thread, so NAMD will report that only 72 cores are being used but this is normal. | ||
| Line 144: | Line 197: | ||
}} | }} | ||
= | === OFI GPU jobs === <!--T:56--> | ||
== OFI GPU jobs == <!--T:56--> | |||
<!--T:57--> | <!--T:57--> | ||
| Line 189: | Line 220: | ||
}} | }} | ||
== | === Verbs-GPU jobs === <!--T:20--> | ||
<!--T:59--> | <!--T:59--> | ||