Configure Remote Simulation Execution

Note

This step is only required if you want to offload heavy simulations to HPC clusters (e.g., SURF Cloud). To use this feature, a cloud/HPC account with access to a cluster with Singularity installed is required. If you do not have access to such a cluster, you can skip this section and run CHORAS locally on your machine.

SSH Key Setup

Generate and configure an SSH key on your machine following the GitHub guide: Generating a new SSH key and adding it to the ssh-agent

In the root directory of the CHORAS repository, open docker-compose.yml. Add the path to your local SSH key directory in volumes of backend:

platform: linux/amd64
build:
  context: . # root of CHORAS
  dockerfile: backend/Dockerfile
ports:
  - "5001:5001"
env_file:
  - .env.api
depends_on:
  - db_service
  - redis
volumes:
  - /var/run/docker.sock:/var/run/docker.sock
  - ./uploads:/app/uploads
  - ./simulation-backend:/app/simulation-backend
  - <absolute-path-to-ssh-directory>/.ssh:/root/.ssh:ro # Add this line

Cloud Configuration Variables

After SSH is configured and you receive your cloud access details (IP address, username, etc.):

1. Go to the backend submodule:

   cd backend
   

2. Open config.py and update the CloudConfig class at the end of the file:

   class CloudConfig:
       """
       Cloud Configuration
       """
       CLOUD_EXECUTOR_HOST = "<cloud-ip>"           # ← Update with your cloud IP
       CLOUD_EXECUTOR_USER = "<cloud-user>"         # ← Update with your username
       CLOUD_EXECUTOR_KEY_PATH = f"{Path.home()}/.ssh/id_ed25519"
       CLOUD_EXECUTOR_DIRECTORY = f"/data/storage/{CLOUD_EXECUTOR_USER}"
   

3. In the CloudConfig class, update:

   • The IP address of the cloud

   • The username

   with the values provided by your cloud/HPC provider (e.g., SURF).

Note

For Cloud/HPC, Singularity should be installed on it.