This repo provides a simple example about deploying multiple ray applications using docker compose
-
couple of ml applications text summarizer and translator that exposes end points for performing summarizing and translation (which are taken from ray documentation)
-
docker file with common dependencies and to dockerize the applications
-
config file used for running the applications
-
docker-compose.yaml that creates a ray cluster with one head and oneworker node(this can be scaled based on requirement and number of applications)
-
client.py to test the service end points
-
docker and docker compose are pre-requisites to run this example
-
checkout the code and go to the root directory
-
create the docker image by running the following command docker buildx build -t ml-apps:latest .
-
once the docker image is built run the following command docker compose up
-
once the service is up one can access the dashboard @ http://localhost:8201
-
once the deployment is ready(one can see the status in dashboard), services can be tested by running client.py or using any http client or postman