-
Notifications
You must be signed in to change notification settings - Fork 13
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
graphcast hub #977
Open
julienchastang
wants to merge
8
commits into
Unidata:master
Choose a base branch
from
julienchastang:tm24f
base: master
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
graphcast hub #977
Changes from all commits
Commits
Show all changes
8 commits
Select commit
Hold shift + click to select a range
137c70e
graphcast hub
julienchastang 5bdaf33
Add some niceties; conda-->mamba
ana-v-espinoza 16dd45a
Modify resource limits
ana-v-espinoza 1b34df1
additional niceties
julienchastang 95df6f6
Revert to standard env before changing
ana-v-espinoza a82fafd
Env: ai-models-graphcast and deps
ana-v-espinoza cf9feb6
Add graphcast dep
ana-v-espinoza 7364b79
earth2mip
julienchastang File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,2 @@ | ||
| envs_dirs: | ||
| - /home/jovyan/additional-envs |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,43 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "cell_type": "markdown", | ||
| "id": "c86cd54f-b73c-4781-b6eb-89c79d3d3b22", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## Acknowledgements\n", | ||
| "\n", | ||
| "Launching this JupyterHub server is the result of a collaboration between several research and academic institutions and their staff. For Jetstream2 and JupyterHub expertise, we thank Andrea Zonca (San Diego Supercomputing Center), Jeremy Fischer, Mike Lowe (Indiana University), the NSF Jetstream2 (`doi:10.1145/3437359.3465565`) team.\n", | ||
| "\n", | ||
| "This work employs the NSF Jetstream2 Cloud at Indiana University through allocation EES220002 from the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS) program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296.\n", | ||
| "\n", | ||
| "Unidata is one of the University Corporation for Atmospheric Research (UCAR)'s Community Programs (UCP), and is funded primarily by the National Science Foundation (AGS-2403649).\n", | ||
| "\n", | ||
| "## To Acknowledge This JupyterHub and the Unidata Science Gateway\n", | ||
| "\n", | ||
| "If you have benefited from the Unidata Science Gateway, please cite `doi:10.5065/688s-2w73`. Additional citation information can be found in this [Citation File Format file](https://raw.githubusercontent.com/Unidata/science-gateway/master/CITATION.cff).\n" | ||
| ] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "Python 3 (ipykernel)", | ||
| "language": "python", | ||
| "name": "python3" | ||
| }, | ||
| "language_info": { | ||
| "codemirror_mode": { | ||
| "name": "ipython", | ||
| "version": 3 | ||
| }, | ||
| "file_extension": ".py", | ||
| "mimetype": "text/x-python", | ||
| "name": "python", | ||
| "nbconvert_exporter": "python", | ||
| "pygments_lexer": "ipython3", | ||
| "version": "3.10.6" | ||
| } | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 5 | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,42 @@ | ||
| # Heavily borrowed from docker-stacks/minimal-notebook/ | ||
| # https://github.com/jupyter/docker-stacks/blob/main/minimal-notebook/Dockerfile | ||
|
|
||
| ARG BASE_CONTAINER=jupyter/tensorflow-notebook | ||
| FROM $BASE_CONTAINER | ||
|
|
||
| ENV DEFAULT_ENV_NAME=tm-fall-2024 EDITOR=nano VISUAL=nano | ||
|
|
||
| LABEL maintainer="Unidata <[email protected]>" | ||
|
|
||
| USER root | ||
|
|
||
| RUN apt-get update && \ | ||
| apt-get install -y --no-install-recommends vim nano curl zip unzip && \ | ||
| apt-get clean && \ | ||
| rm -rf /var/lib/apt/lists/* | ||
|
|
||
| USER $NB_UID | ||
|
|
||
| ADD environment.yml /tmp | ||
|
|
||
| RUN mamba install --quiet --yes \ | ||
| 'conda-forge::nb_conda_kernels' \ | ||
| 'conda-forge::jupyterlab-git' \ | ||
| 'conda-forge::ipywidgets' && \ | ||
| mamba env update --name $DEFAULT_ENV_NAME -f /tmp/environment.yml && \ | ||
| pip install --no-cache-dir nbgitpuller && \ | ||
| mamba clean --all -f -y && \ | ||
| jupyter lab clean -y && \ | ||
| npm cache clean --force && \ | ||
| rm -rf /home/$NB_USER/.cache/yarn && \ | ||
| rm -rf /home/$NB_USER/.node-gyp && \ | ||
| fix-permissions $CONDA_DIR && \ | ||
| fix-permissions /home/$NB_USER | ||
|
|
||
| COPY GPU_sanity_check.ipynb Acknowledgements.ipynb \ | ||
| default_kernel.py .condarc / | ||
|
|
||
| ARG JUPYTER_SETTINGS_DIR=/opt/conda/share/jupyter/lab/settings/ | ||
| COPY overrides.json $JUPYTER_SETTINGS_DIR | ||
|
|
||
| USER $NB_UID | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,287 @@ | ||
| { | ||
| "cells": [ | ||
| { | ||
| "cell_type": "markdown", | ||
| "id": "f1b605ba-8d44-4654-be99-df2d39289c36", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "## GPU JHub Testing Notebook" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "markdown", | ||
| "id": "25cacd08-18b8-4991-966e-7e49aa44192a", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "Notebook used for first pass testing of the environment and GPU access. Here are the various JS2 GPU instance [flavors](https://docs.jetstream-cloud.org/general/instance-flavors/#jetstream2-gpu)." | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "markdown", | ||
| "id": "8f0798c9-ee1d-4157-a4be-deedd90bd9a5", | ||
| "metadata": {}, | ||
| "source": [ | ||
| "Note: this also tests PyTorch install, as of Novembeerr 2024, I hope to not use tensorflow for work at UCAR / Unidata. This entire notebook should run without any errors. " | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 1, | ||
| "id": "831b1a5d-488d-476c-8050-4f18cd635c0c", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "import psutil\n", | ||
| "import platform\n", | ||
| "import sys\n", | ||
| "\n", | ||
| "import torch\n", | ||
| "import platform" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 2, | ||
| "id": "50c795b6-34a9-4c2b-ab71-67c5ea087fa2", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def get_simple_system_info():\n", | ||
| " # Memory info\n", | ||
| " memory = psutil.virtual_memory()\n", | ||
| " ram_gb = memory.total / (1024 ** 3) # Convert to GB\n", | ||
| " ram_used_gb = memory.used / (1024 ** 3)\n", | ||
| " \n", | ||
| " # CPU info\n", | ||
| " cpu_cores = psutil.cpu_count()\n", | ||
| " cpu_usage = psutil.cpu_percent(interval=1)\n", | ||
| " \n", | ||
| " print(f\"Python Version: {platform.python_version()}\")\n", | ||
| " print(f\"\\nCPU:\")\n", | ||
| " print(f\"- Cores: {cpu_cores}\")\n", | ||
| " print(f\"- Current Usage: {cpu_usage}%\")\n", | ||
| " print(f\"\\nRAM:\")\n", | ||
| " print(f\"- Total: {ram_gb:.1f} GB\")\n", | ||
| " print(f\"- Used: {ram_used_gb:.1f} GB\")\n", | ||
| " print(f\"- Usage: {memory.percent}%\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 3, | ||
| "id": "7ce932bc-9406-4841-91ca-371e3c768980", | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "Python Version: 3.10.15\n", | ||
| "\n", | ||
| "CPU:\n", | ||
| "- Cores: 8\n", | ||
| "- Current Usage: 1.0%\n", | ||
| "\n", | ||
| "RAM:\n", | ||
| "- Total: 29.4 GB\n", | ||
| "- Used: 1.4 GB\n", | ||
| "- Usage: 6.1%\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "get_simple_system_info()" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 4, | ||
| "id": "9aa829ef-1093-4d6d-9052-a86a1a8647fc", | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "Thu Nov 21 16:36:04 2024 \n", | ||
| "+---------------------------------------------------------------------------------------+\n", | ||
| "| NVIDIA-SMI 535.183.06 Driver Version: 535.183.06 CUDA Version: 12.2 |\n", | ||
| "|-----------------------------------------+----------------------+----------------------+\n", | ||
| "| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |\n", | ||
| "| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |\n", | ||
| "| | | MIG M. |\n", | ||
| "|=========================================+======================+======================|\n", | ||
| "| 0 GRID A100X-10C On | 00000000:04:00.0 Off | 0 |\n", | ||
| "| N/A N/A P0 N/A / N/A | 0MiB / 10240MiB | 0% Default |\n", | ||
| "| | | Disabled |\n", | ||
| "+-----------------------------------------+----------------------+----------------------+\n", | ||
| " \n", | ||
| "+---------------------------------------------------------------------------------------+\n", | ||
| "| Processes: |\n", | ||
| "| GPU GI CI PID Type Process name GPU Memory |\n", | ||
| "| ID ID Usage |\n", | ||
| "|=======================================================================================|\n", | ||
| "| No running processes found |\n", | ||
| "+---------------------------------------------------------------------------------------+\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "!nvidia-smi" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 5, | ||
| "id": "0bb2ce0f-afba-4693-8072-bccb92dca0bf", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def get_pytorch_info():\n", | ||
| " print(\"PyTorch System Information\")\n", | ||
| " print(\"-\" * 30)\n", | ||
| " \n", | ||
| " # PyTorch version\n", | ||
| " print(f\"PyTorch Version: {torch.__version__}\")\n", | ||
| " \n", | ||
| " # CUDA availability\n", | ||
| " print(f\"\\nCUDA Available: {torch.cuda.is_available()}\")\n", | ||
| " \n", | ||
| " if torch.cuda.is_available():\n", | ||
| " # Current device information\n", | ||
| " current_device = torch.cuda.current_device()\n", | ||
| " print(f\"Current CUDA Device: {current_device}\")\n", | ||
| " \n", | ||
| " # Device name\n", | ||
| " print(f\"Device Name: {torch.cuda.get_device_name(current_device)}\")\n", | ||
| " \n", | ||
| " # CUDA version\n", | ||
| " print(f\"CUDA Version: {torch.version.cuda}\")\n", | ||
| " \n", | ||
| " # Number of CUDA devices\n", | ||
| " print(f\"Device Count: {torch.cuda.device_count()}\")\n", | ||
| " \n", | ||
| " # Memory information\n", | ||
| " print(\"\\nGPU Memory Information:\")\n", | ||
| " print(f\"- Total: {torch.cuda.get_device_properties(current_device).total_memory / 1024**3:.2f} GB\")\n", | ||
| " print(f\"- Allocated: {torch.cuda.memory_allocated(current_device) / 1024**3:.2f} GB\")\n", | ||
| " print(f\"- Cached: {torch.cuda.memory_reserved(current_device) / 1024**3:.2f} GB\")\n", | ||
| " \n", | ||
| " # Architecture information\n", | ||
| " device_props = torch.cuda.get_device_properties(current_device)\n", | ||
| " print(f\"\\nGPU Architecture:\")\n", | ||
| " print(f\"- GPU Compute Capability: {device_props.major}.{device_props.minor}\")\n", | ||
| " print(f\"- Multi Processors: {device_props.multi_processor_count}\")\n", | ||
| " else:\n", | ||
| " print(\"\\nNo CUDA GPU available. PyTorch will run on CPU only.\")\n", | ||
| " print(f\"CPU Architecture: {platform.machine()}\")\n", | ||
| " print(f\"CPU Type: {platform.processor()}\")" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 6, | ||
| "id": "ea2e96f2-37fe-49a0-8cc8-d32a3b666a0a", | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "name": "stdout", | ||
| "output_type": "stream", | ||
| "text": [ | ||
| "PyTorch System Information\n", | ||
| "------------------------------\n", | ||
| "PyTorch Version: 2.5.1+cu124\n", | ||
| "\n", | ||
| "CUDA Available: True\n", | ||
| "Current CUDA Device: 0\n", | ||
| "Device Name: GRID A100X-10C\n", | ||
| "CUDA Version: 12.4\n", | ||
| "Device Count: 1\n", | ||
| "\n", | ||
| "GPU Memory Information:\n", | ||
| "- Total: 10.00 GB\n", | ||
| "- Allocated: 0.00 GB\n", | ||
| "- Cached: 0.00 GB\n", | ||
| "\n", | ||
| "GPU Architecture:\n", | ||
| "- GPU Compute Capability: 8.0\n", | ||
| "- Multi Processors: 108\n" | ||
| ] | ||
| } | ||
| ], | ||
| "source": [ | ||
| "get_pytorch_info()" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 7, | ||
| "id": "b3f8dad5-6c3c-4a64-bc31-b9ef5d5894fb", | ||
| "metadata": {}, | ||
| "outputs": [], | ||
| "source": [ | ||
| "def get_instance_type():\n", | ||
| " cpu_count = psutil.cpu_count()\n", | ||
| " ram_gb = psutil.virtual_memory().total / (1024**3)\n", | ||
| " gpu_ram = 0\n", | ||
| " \n", | ||
| " if torch.cuda.is_available():\n", | ||
| " current_device = torch.cuda.current_device()\n", | ||
| " gpu_ram = torch.cuda.get_device_properties(current_device).total_memory / (1024**3)\n", | ||
| " \n", | ||
| " if cpu_count == 4 and 13 <= ram_gb <= 17 and 7 <= gpu_ram <= 9:\n", | ||
| " return \"g3.small\"\n", | ||
| " elif cpu_count == 8 and 28 <= ram_gb <= 32 and 9 <= gpu_ram <= 11:\n", | ||
| " return \"g3.medium\"\n", | ||
| " elif cpu_count == 16 and 58 <= ram_gb <= 62 and 19 <= gpu_ram <= 21:\n", | ||
| " return \"g3.large\"\n", | ||
| " elif cpu_count == 32 and 123 <= ram_gb <= 127 and 39 <= gpu_ram <= 41:\n", | ||
| " return \"g3.xl\"\n", | ||
| " else:\n", | ||
| " return \"custom\"" | ||
| ] | ||
| }, | ||
| { | ||
| "cell_type": "code", | ||
| "execution_count": 8, | ||
| "id": "ef30c0c2-bf93-4c7a-97e2-c80f7c503f20", | ||
| "metadata": {}, | ||
| "outputs": [ | ||
| { | ||
| "data": { | ||
| "text/plain": [ | ||
| "'g3.medium'" | ||
| ] | ||
| }, | ||
| "execution_count": 8, | ||
| "metadata": {}, | ||
| "output_type": "execute_result" | ||
| } | ||
| ], | ||
| "source": [ | ||
| "get_instance_type()" | ||
| ] | ||
| } | ||
| ], | ||
| "metadata": { | ||
| "kernelspec": { | ||
| "display_name": "Python [conda env:tm-fall-2024]", | ||
| "language": "python", | ||
| "name": "conda-env-tm-fall-2024-py" | ||
| }, | ||
| "language_info": { | ||
| "codemirror_mode": { | ||
| "name": "ipython", | ||
| "version": 3 | ||
| }, | ||
| "file_extension": ".py", | ||
| "mimetype": "text/x-python", | ||
| "name": "python", | ||
| "nbconvert_exporter": "python", | ||
| "pygments_lexer": "ipython3", | ||
| "version": "3.10.15" | ||
| } | ||
| }, | ||
| "nbformat": 4, | ||
| "nbformat_minor": 5 | ||
| } |
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Why the switch from minimal to the tf image?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It is the only way I can get the GPU to work.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What are you trying? The image that's live right now uses the minimal-notebook and thomas and I can access the GPU just fine.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
How did you get that to work beyond the usual, i.e., the newly defined
environment.ymlandjupyterhub_gpu.yaml. Did you have to install anything special CUDA/GPU-wise?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
My understanding is that the current running environment was modified "in-place" for faster experimentation so not everything may have been captured in the
Dockerfile,environment.yml, etc. I guess that is what I am asking.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You can find the most recent Dockerfile/environment combo in the "ana" tmux session on the docker-gpu machine. Your most recent commit has the correct environment file, however using the minimal-notebook as the base still works.
To get the GPU to work there is nothing more needed past the "normal" things--i.e. requesting the additional resource via
jupyterhub_gpu.yaml.I've found that the important parts of getting these things to work are making sure the packages you install (via
condaorpip) are compatible with that shown when doing annvidia-smion JS2.For example, in this case I tell
pipto look fortorchand associated packages from the cuda 12.1 index with this line:- --extra-index-url https://download.pytorch.org/whl/cu121We can get together to discuss this in-person after the holidays?