Dockerized setup for sonar

Dockerfile

@@ -0,0 +1,32 @@
+# Use the latest LTS (Long-Term Support) version of Ubuntu
+FROM ubuntu:latest
+
+# Set working directory to /sonar
+WORKDIR /sonar
+
+#Install base utilities and Python with required dependencies
+RUN apt-get update && \
+    apt-get install -y software-properties-common && \
+    add-apt-repository ppa:deadsnakes/ppa && \
+    apt-get update && \
+    apt-get install -y \
+    build-essential \
+    wget \
+    python3.11 \
+    python3-pip \
+    python3-venv \
+    openmpi-bin \
+    libopenmpi-dev \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements.txt to the working directory
+COPY requirements.txt .
+
+# Create a virtual environment and install dependencies
+RUN python3 -m venv env && \
+    . env/bin/activate && \
+    env/bin/pip install -r requirements.txt
+
+# To activate the virtual environment by default, use ENTRYPOINT
+ENTRYPOINT ["/bin/bash", "-c", "source /sonar/env/bin/activate && exec \"$@\"", "--"]

docker_run.sh

@@ -0,0 +1 @@
+docker run -it --rm -v $(pwd)/src:/sonar/src sonar_image /bin/bash -c "cd /sonar/src && mpirun --allow-run-as-root -np 4 -host localhost:11 ../env/bin/python main.py"

requirements.txt

@@ -23,18 +23,18 @@ mpi4py==3.1.6
 mpmath==1.3.0
 networkx==3.3
 numpy==1.26.4
-nvidia-cublas-cu12==12.1.3.1
-nvidia-cuda-cupti-cu12==12.1.105
-nvidia-cuda-nvrtc-cu12==12.1.105
-nvidia-cuda-runtime-cu12==12.1.105
-nvidia-cudnn-cu12==8.9.2.26
-nvidia-cufft-cu12==11.0.2.54
-nvidia-curand-cu12==10.3.2.106
-nvidia-cusolver-cu12==11.4.5.107
-nvidia-cusparse-cu12==12.1.0.106
-nvidia-nccl-cu12==2.20.5
-nvidia-nvjitlink-cu12==12.5.40
-nvidia-nvtx-cu12==12.1.105
+nvidia-cublas-cu12
+nvidia-cuda-cupti-cu12>=12.1.105
+nvidia-cuda-nvrtc-cu12>=12.1.105
+nvidia-cuda-runtime-cu12>=12.1.105
+nvidia-cudnn-cu12>=8.9.2.26
+nvidia-cufft-cu12>=11.0.2.54
+nvidia-curand-cu12>=10.3.2.106
+nvidia-cusolver-cu12>=11.4.5.107
+nvidia-cusparse-cu12>=12.1.0.106
+nvidia-nccl-cu12>=2.20.5
+nvidia-nvjitlink-cu12>=12.5.40
+nvidia-nvtx-cu12>=12.1.105
 ogb==1.3.6
 outdated==0.2.2
 packaging==24.0
@@ -57,7 +57,6 @@ tifffile==2024.5.22
 torch==2.3.0
 torchvision==0.18.0
 tqdm==4.66.4
-triton==2.3.0
 typing_extensions==4.12.0
 tzdata==2024.1
 urllib3==2.2.1

src/README.md

@@ -1,3 +1,45 @@
+## Setting Up the Project with Docker
+
+### Prerequisites
+
+Ensure you have Docker installed on your machine. You can download it from [Docker's official website](https://www.docker.com/get-started).
+
+### Building the Docker Image
+
+1. Clone the repository:
+    ```sh
+    git clone https://github.com/aidecentralized/sonar
+    cd sonar
+    ```
+
+2. Build the Docker image:
+    ```sh
+    docker build -t sonar_image:latest .
+    ```
+
+### Running the Container
+
+We provide a `docker_run.sh` script to simplify running the Docker container.
+
+1. Ensure the script has execution permissions:
+    ```sh
+    chmod +x docker_run.sh
+    ```
+
+2. Run the script:
+- Using './':
+    ```sh
+    ./docker_run.sh
+    ```
+- Using 'bash':
+    ```sh
+    bash docker_run.sh
+    ```
+
+The `docker_run.sh` script will handle the necessary Docker commands to start the container with the appropriate settings.
+
+Note: If you have used a different Image name while building Docker Image, Please update the name accordingly in `docker_run.sh` script
+
 ### Running the code
 Let's say you want to run the model training of 3 nodes on a machine. That means there will be 4 nodes in total because there is 1 more node in addition to the clients --- server.
 The whole point of this project is to eventually transition to a distributed system where each node can be a separate machine and a server is simply another node. But for now, this is how things are done.