This brief introduction shows you how to deploy Feast using Docker Compose. Docker Compose enables you to explore the functionality provided by Feast while requiring only minimal infrastructure. It includes a built in Jupyter Notebook Server that is preloaded with PySpark and the Feast Python SDK, as well as Feast example notebooks to get you started.
- Install Docker Compose
- Optional dependencies:
- a GCP service account that has access to Google Cloud Storage
Before starting, clone the latest stable version of Feast from the Feast repository, and configure.
git clone https://github.com/feast-dev/feast.git
cd feast/infra/docker-compose
cp .env.sample .env
Start the Feast services. Ensure that the following ports are free on the host machines: 6565, 6566, 8888, 9094, 5432. Alternatively, change port mapping to use a different port on the host.
docker-compose up -d{% hint style="info" %} Docker Compose takes a minute to launch:
- While the deployment is initiating, Feast Online Serving container may restart. Once Feast Core has launched, any container restarts should be auto-corrected.
- If container restarts continue for more than 10 minutes, check the Docker Compose log for any errors that prevent Feast Core from starting successfully. {% endhint %}
You can now connect to the bundled Jupyter Notebook Server at localhost:8888 and follow the example notebooks.
{% embed url="http://localhost:8888/tree?" caption="" %}
By default, the example Jupyter notebook does not require any GCP dependencies. If you modify the example to require GCP services (for example Google Cloud Storage), you need to establish a service account that is associated with the Jupyter notebook.
{% hint style="warning" %} Confirm that the GCP service account has sufficient privileges to access the required GCP services. {% endhint %}
After you create the service account, download the associated JSON key file and copy the file to the path configured in .env under GCP_SERVICE_ACCOUNT .