Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revisionBoth sides next revision | ||
pandoc:introduction-to-vsc:09_special_hardware:accelerators [2020/10/20 08:09] – Pandoc Auto-commit pandoc | pandoc:introduction-to-vsc:09_special_hardware:accelerators [2024/06/06 15:40] – [TOP500 List June 2024] goldenberg | ||
---|---|---|---|
Line 1: | Line 1: | ||
+ | ====== GPUs available & how to use it ====== | ||
- | ====== Special hardware (GPUs, binfs) available & how to use it ====== | + | ===== TOP500 List June 2024 ===== |
- | | + | ^ Rank^Nation |
+ | | 1.|{{.: | ||
+ | | 2.|{{.: | ||
+ | | 3.|{{.: | ||
+ | | 4.|{{.: | ||
+ | | 5.| | LUMI | 379 PFLOPs/ | ||
+ | | 6.|{{.: | ||
+ | | 7.|{{.: | ||
+ | | 8.| | MareNostrum 5 ACC | 175 PFLOPs/ | ||
+ | | 9.|{{.: | ||
+ | | 10.|{{.: | ||
- | ====== TOP500 List June 2020 ====== | + | ===== Components on VSC-5 ===== |
+ | ^Model ^# | ||
+ | |19x GeForce RTX-2080Ti n375-[001-019] - only in a special project | ||
+ | |{{: | ||
+ | |45x2 nVidia A40 n306[6, | ||
+ | |{{ : | ||
+ | |62x2 nVidia A100-40GB n307[1-4]-[001-015] | ||
+ | |{{ : | ||
- | < | ||
- | < | ||
- | <!--for nations flags see https:// | ||
- | </ | ||
- | ^ Rank^Nation | ||
- | | 1.|{{..: | ||
- | | 2.|{{..: | ||
- | | 3.|{{..: | ||
- | | 4.|{{..: | ||
- | | 5.|{{..: | ||
- | | 6.|{{..: | ||
- | | 7.|{{..: | ||
- | | 8.|{{..: | ||
- | | 9.|{{..: | ||
- | | | ||
- | + | ==== Working on GPU nodes Interactively | |
- | < | + | |
- | < | + | |
- | </ | + | |
- | ====== Components on VSC-3 ====== | + | |
- | + | ||
- | ^Model | + | |
- | |< | + | |
- | |{{..: | + | |
- | |< | + | |
- | |{{..: | + | |
- | |< | + | |
- | |{{..: | + | |
- | + | ||
- | + | ||
- | < | + | |
- | < | + | |
- | </ | + | |
- | ====== Working on GPU nodes ====== | + | |
**Interactive mode** | **Interactive mode** | ||
< | < | ||
- | 1. VSC-3 > salloc -N 1 -p gpu_gtx1080single | + | 1. VSC-5 > salloc -N 1 -p zen2_0256_a40x2 |
- | 2. VSC-3 > squeue -u $USER | + | 2. VSC-5 > squeue -u $USER |
- | 3. VSC-3 > srun -n 1 hostname | + | 3. VSC-5 > srun -n 1 hostname |
- | 4. VSC-3 > ssh n372-012 (...or whatever else node had been assigned) | + | 4. VSC-5 > ssh n3066-012 (...or whatever else node had been assigned) |
- | 5. VSC-3 > module load cuda/ | + | 5. VSC-5 > module load cuda/ |
cd ~/ | cd ~/ | ||
nvcc ./ | nvcc ./ | ||
Line 63: | Line 48: | ||
./a.out | ./a.out | ||
- | 6. VSC-3 > nvidia-smi | + | 6. VSC-5 > nvidia-smi |
- | 7. VSC-3 > / | + | 7. VSC-5 > / |
</ | </ | ||
- | < | ||
- | < | ||
- | </ | ||
- | ====== Working on GPU nodes cont. ====== | ||
- | **SLURM submission** | + | ===== Working on GPU using SLURM ===== |
+ | |||
+ | **SLURM submission** gpu_test.scrpt | ||
<code bash> | <code bash> | ||
Line 79: | Line 62: | ||
# usage: sbatch ./ | # usage: sbatch ./ | ||
# | # | ||
- | #SBATCH -J gtx1080 | + | #SBATCH -J A40 |
- | #SBATCH -N 1 | + | #SBATCH -N 1 #use -N only if you use both GPUs on the nodes, otherwise leave this line out |
- | #SBATCH --partition | + | #SBATCH --partition |
- | #SBATCH --qos gpu_gtx1080single | + | #SBATCH --qos zen2_0256_a40x2 |
+ | #SBATCH --gres=gpu: | ||
module purge | module purge | ||
Line 90: | Line 74: | ||
/ | / | ||
</ | </ | ||
- | < | ||
- | |||
- | < | ||
- | < | ||
- | </ | ||
- | ====== Working on binf nodes ====== | ||
- | |||
- | **Interactive mode** | ||
- | |||
- | < | ||
- | 1. VSC-3 > salloc -N 1 -p binf --qos normal_binf -C binf -L intel@vsc | ||
- | (... add | ||
- | |||
- | 2. VSC-3 > squeue -u $USER | ||
- | |||
- | 3. VSC-3 > srun -n 4 hostname | ||
- | |||
- | 4. VSC-3 > ssh binf-11 | ||
- | |||
- | 5. VSC-3 > module purge | ||
- | |||
- | 6. VSC-3 > module load intel/ | ||
- | cd examples/ | ||
- | icc -xHost -qopenmp sample.c | ||
- | export OMP_NUM_THREADS=8 | ||
- | ./a.out | ||
- | </ | ||
- | < | ||
- | < | ||
- | </ | ||
- | ====== Working on binf nodes cont. ====== | ||
- | |||
- | **SLURM submission** [[examples/ | ||
- | |||
- | <code bash> | ||
- | #!/bin/bash | ||
- | # | ||
- | # usage: sbatch ./ | ||
- | # | ||
- | #SBATCH -J gmxbinfs | ||
- | #SBATCH -N 2 | ||
- | #SBATCH --partition binf | ||
- | #SBATCH --qos normal_binf | ||
- | #SBATCH -C binf | ||
- | #SBATCH --ntasks-per-node 24 | ||
- | #SBATCH --ntasks-per-core 1 | ||
- | |||
- | module purge | ||
- | module load intel/ | ||
- | |||
- | export I_MPI_PIN=1 | ||
- | export I_MPI_PIN_PROCESSOR_LIST=0-23 | ||
- | export I_MPI_FABRICS=shm: | ||
- | export I_MPI_TMI_PROVIDER=psm2 | ||
- | export OMP_NUM_THREADS=1 | ||
- | export MDRUN_ARGS=" | ||
- | |||
- | mpirun -np $SLURM_NTASKS gmx_mpi mdrun ${MDRUN_ARGS} | ||
- | </ | ||
- | < | ||
- | < | ||
- | </ | ||
- | ====== Real-World Example, AMBER-16 ====== | ||
- | ^ Performance^Power Efficiency | ||
- | | {{..: | ||
+ | ===== Real-World Example, AMBER-16 ===== | ||
+ | ^ | ||
+ | | {{.: | ||