Features/1716 update docker image

docker/AMD/Dockerfile.release

@@ -0,0 +1,30 @@
+################################################################################################
+# Development Environment
+# Pulls Pytorch image and installs heat + dependencies
+################################################################################################
+
+# Build Arguments
+ARG DEFAULT=latest
+ARG HEAT_VERSION=1.5.x
+ARG ROCM_VERSION=6.2
+ARG AMDGPU_VERSION=6.2
+
+ARG VERSION_STRING=rocm6.3_ubuntu22.04_py3.10_pytorch_release_2.3.0
+
+
+# This is the base image for the release image
+FROM rocm/pytorch:${DEFAULT} AS base
+COPY ./tzdata.seed /tmp/tzdata.seed
+RUN debconf-set-selections /tmp/tzdata.seed
+
+# Starts a new stage for the release image
+FROM base AS release-install
+# RUN /usr/bin/python3 -m pip install --upgrade pip
+RUN pip install --upgrade pip
+RUN pip install mpi4py --no-binary :all:
+# RUN echo ${HEAT_VERSION}
+# RUN if [[ ${HEAT_VERSION} =~ ^([1-9]\d*|0)(\.(([1-9]\d*)|0)){2}$ ]]; then \
+#        pip install heat[hdf5,netcdf]==${HEAT_VERSION}; \
+#    else \
+#        pip install heat[hdf5,netcdf]; \
+#    fi

docker/AMD/Dockerfile.source

@@ -0,0 +1,18 @@
+########################################################################################
+# Development Environment
+# Pulls Pytorch image and installs heat + dependencies
+########################################################################################
+
+ARG PYTORCH_IMG=24.10-py3
+ARG HEAT_BRANCH=main
+
+FROM nvcr.io/nvidia/pytorch:${PYTORCH_IMG} AS base
+COPY ./tzdata.seed /tmp/tzdata.seed
+RUN debconf-set-selections /tmp/tzdata.seed
+RUN apt update && DEBIAN_FRONTEND=noninteractive apt install -y build-essential openssh-client python3-dev git && apt clean && rm -rf /var/lib/apt/lists/*
+
+FROM base AS source-install
+ARG HEAT_BRANCH
+RUN pip install --upgrade pip
+RUN git clone -b ${HEAT_BRANCH} https://github.com/helmholtz-analytics/heat.git
+RUN pip install mpi4py --no-binary :all: && pushd heat && pip install .[hdf5,netcdf] && popd && rm -rf heat

docker/AMD/README.md

@@ -0,0 +1,19 @@
+# Not functional yet
+
+
+### AMD/ROCM Specific Information
+
+Template found in: [ROCm Docker Github Link](github.com/ROCm/ROCm-docker/blob/master/rocm-terminal/Dockerile)
+
+Check out the following [installation guide](https://github.com/ROCm/ROCm-docker/blob/master/quick-start.md)
+
+### AMD Docker file sources
+#### Basic Image
+[Radeon Repository](https://repo.radeon.com/rocm/manylinux/)
+
+#### Pytorch Image
+[ROCm docker hub](https://hub.docker.com/r/rocm/pytorch)
+
+
+#### General Things I realized
+You might run into an error that no space is left on your device when trying to build the pytorch container. Within Docker Desktop you can specify a limit for file sizes. Since containers can take up to 100GB, it might take some free storage.

docker/AMD/compose.yml

@@ -0,0 +1,7 @@
+# version: '3.8'
+
+services:
+  rocm-dev:
+    image: rocm/dev-ubuntu-24.04
+    container_name: rocm_dev_container
+    restart: unless-stopped

docker/AMD/tzdata.seed

@@ -0,0 +1,2 @@
+tzdata tzdata/Areas select Europe
+tzdata tzdata/Zones/Europe select Berlin

docker/Dockerfile.release

@@ -1,5 +1,12 @@
-ARG HEAT_VERSION=latest
-ARG PYTORCH_IMG=23.05-py3
+####################################################################################################
+# NVIDIA Docker Release File
+# Check out available tags at:  https://catalog.ngc.nvidia.com/orgs/nvidia/containers/pytorch/tags
+# Check out the release notes:  https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-25-02.html
+####################################################################################################
+
+#ARG PYTORCH_IMG=24.10-py3
+ARG PYTORCH_IMG=25.02-py3
+ARG HEAT_VERSION=1.5.1
 
 FROM nvcr.io/nvidia/pytorch:${PYTORCH_IMG} AS base
 COPY ./tzdata.seed /tmp/tzdata.seed

docker/Dockerfile.source

@@ -1,4 +1,9 @@
-ARG PYTORCH_IMG=23.05-py3
+####################################################################################################
+# NVIDIA Docker Source File
+# Development Environment
+####################################################################################################
+
+ARG PYTORCH_IMG=24.10-py3
 ARG HEAT_BRANCH=main
 
 FROM nvcr.io/nvidia/pytorch:${PYTORCH_IMG} AS base

docker/README.md

@@ -97,3 +97,44 @@ srun --mpi="pmi2" apptainer exec --nv heat_1.2.0_torch.11_cuda11.5_py3.9.sif bas
 ## Scripts
 
 The scripts folder has a small collection of helper scripts to automate certain tasks, primarly meant for heat developers. Explanations are given at the top of the script.
+
+## Useful commands for first-time docker users
+
+> docker run -it [docker container id]
+Runs the container in interactive mode (Opens a terminal)
+
+> docker ps -a
+Lists all active containers
+
+> docker system prune
+Closes all currently running containers and frees up the resources
+
+> docker system prune --all --force
+Frees up all space taken up by docker images, even stopped ones
+
+> docker images
+Lists all docker images
+
+> docker tag <old_image_name>:<old_image_tag> <new_image_name>:<new_image_tag>
+Rename the docker container (Needed to upload)
+
+## How to download a pre-built image from the container registry
+The github container registry (ghcr.io) contains different docker versions of heat / pytorch / cuda / rocm.
+
+> docker pull ghcr.io/NAMESPACE/IMAGE_NAME
+
+For further info refer to the [Github documentation on package registries](https://docs.github.com/en/packages/working-with-a-github-packages-registry/working-with-the-container-registry)
+
+## How to push a new image to ghcr.io
+
+1. Make sure you have a github access token set up in the CLI
+> [Authenticating to the container registry](https://docs.github.com/en/packages/working-with-a-github-packages-registry/working-with-the-container-registry#authenticating-to-the-container-registry)
+
+> docker login ghcr.io -u USERNAME -p GITHUB_PERSONAL_ACCESS_TOKEN
+
+2. Rename the local image in the following format:
+> docker tag current:name ghcr.io/helmholtz-analytics/heat:1.X.X-torchX.X_cudaXX.X_py3.XX
+1. Upload the image via:
+> docker push ghcr.io/helmholtz-analytics/heat:1.X.X-torchX.X_cudaXX.X_py3.XX
+
+For further info refer to the [Github documentation on package registries](https://docs.github.com/en/packages/working-with-a-github-packages-registry/working-with-the-container-registry)