Delta (NCSA)
Platform user guide
General description
Resource manager -
SLURM
Launch methods (per platform ID)
ncsa.delta*
-SRUN
Configuration per node (per platform ID)
ncsa.delta
(132 nodes)128 CPU cores, each core has 1 thread
256 GiB of memory
ncsa.delta_gpu_a40
(100 nodes)64 CPU cores, each core has 1 thread
4 GPUs (NVIDIA A40)
256 GiB of memory
ncsa.delta_gpu_a100_4way
(100 nodes)64 CPU cores, each core has 1 thread
4 GPUs (NVIDIA A100)
256 GiB of memory
ncsa.delta_gpu_a100_8way
(6 nodes)128 CPU cores, each core has 1 thread
8 GPUs (NVIDIA A100)
2,048 GiB of memory
ncsa.delta_gpu_mi100
(1 node)128 CPU cores, each core has 1 thread
8 GPUs (AMD MI100)
2,048 GiB of memory
Note
Use the accounts
command to list the accounts available for charging
(Local Account Charging).
Note
Use the quota
command to view your use of the file systems and use by
your projects (Quota Usage).
Setup execution environment
Python virtual environment
Create a virtual environment with venv
:
export PYTHONNOUSERSITE=True
module load python
python3 -m venv ve.rp
source ve.rp/bin/activate
Install RADICAL-Pilot after activating a corresponding virtual environment:
pip install radical.pilot
Note
Polaris does not provide virtual environments with conda
.
Launching script example
Launching script (e.g., rp_launcher.sh
) for the RADICAL-Pilot application
includes setup processes to activate a certain execution environment and
launching command for the application itself.
#!/bin/sh
# - pre run -
module load python
source ve.rp/bin/activate
export RADICAL_PROFILE=TRUE
# for debugging purposes
export RADICAL_LOG_LVL=DEBUG
# - run -
python <rp_application>
Execute launching script as ./rp_launcher.sh
or run it in the background:
nohup ./rp_launcher.sh > OUTPUT 2>&1 </dev/null &
# check the status of the script running:
# jobs -l
Note
If you find any inaccuracy in this description, please, report back to us by opening a ticket.