Example configurations for using pyhf with HTCondor inspired by the Center for High Throughput Computing examples.
These examples assume that you want to use GPU resources to take advantage of hardware acceleration and so focus on using the pyhf Docker base images built on the NVIDIA CUDA enabled images for runtime use with the the NVIDIA Container Toolkit.
- Make sure that you have the
nvidia-container-toolkitinstalled on the host machine - Check the list of available tags on Docker Hub to find the tag you want
- Use
docker pullto pull down the image corresponding to the tag
Example:
docker pull pyhf/cuda:0.7.2-jax-cuda-11.6.0-cudnn8
To check that NVIDIA GPUS are being properly detected run
docker run --rm --gpus all pyhf/cuda:0.7.2-jax-cuda-11.6.0-cudnn8 'nvidia-smi'
and check if the nvidia-smi output appears correctly.
To run (interactively) using GPUs on the host machine:
docker run --rm -ti --gpus all pyhf/cuda:0.7.2-jax-cuda-11.6.0-cudnn8