podman_local

Feast example using Podman and Podman Compose

This guide explains how to deploy Feast remote server components using Podman Compose locally and run an example using the client.

Prerequisites

1. Podman: Podman installation guide.
2. Podman Compose: Podman Compose Installation guide.
3. Python 3.9+ environment
4. Feast CLI

Setup

1. Feast Project Setup

• The project feature_repo already created using feast init command

2. Run the Podman Compose File

• Use the docker-compose.yml file to install and run the Feast feature servers (online, offline, and registry) on podman. The docker-compose file uses the quay.io/feastdev/feature-server:latest image. Each respective service has specific port mappings and maps the volume from the ./feature_repo configuration.

• To start the feature servers, run the following command:

  podman-compose up -d

• This will launch the necessary containers for online, offline, and registry feature servers.

3. Verify the Installation

• Use the podman ps command to verify that the containers are running:

  podman ps

  Example output:

  CONTAINER ID  IMAGE                                     COMMAND               CREATED        STATUS        PORTS                   NAMES
  61442d6d6ef3  quay.io/feastdev/feature-server:latest  feast -c /feature...  2 minutes ago  Up 2 minutes  0.0.0.0:6566->6566/tcp  online-feature-server
  1274c21716a6  quay.io/feastdev/feature-server:latest  feast -c /feature...  2 minutes ago  Up 2 minutes  0.0.0.0:8815->8815/tcp  offline-feature-server
  4e38ca8c39db  quay.io/feastdev/feature-server:latest  feast -c /feature...  2 minutes ago  Up 2 minutes  0.0.0.0:6570->6570/tcp  registry-feature-server
  

• Alternatively, you can verify the running containers through Podman Desktop:

4. Run Feast Apply

• To apply the feature store definitions to the remote registry, run the following command:

  podman exec registry-feature-server feast -c /feature_repo apply

5. Run Client Examples

• The client folder contains example client-side configurations and code:
  • feature_store.yaml: Configuration for the feature store.
  • test.py: Example Python script to interact with the Feast server.

6. Cleanup

• To stop and remove the running containers, run the following command:

  podman-compose down

• This will stop all the feature server containers and clean up the environment.