Custom Services
Any custom or 3rd party AI/ML applications that administrators wish to publish as service profiles can be made available for consumption by end users with a "1-click" experience.
Rafay provides a number of services that administrators can make available as custom services to their end users.
- Inference as a Service
- Ray as a Service
- MLOps based on Kubeflow
- Fine Tuning as a Service
New Custom Service¶
To create a new custom service,
- Select Custom Service from the menu on the left of the console
- Click on add/new custom service
- Select a suitable profile that suits your requirements
- Provide a name for the service (optional description, display name)
- Select the compute instance from the drop down you would like to deploy the custom service to
- Click on save and launch the custom service.
Depending on the complexity of the custom service app, it can take a few minutes for all the components to be deployed, become operational and usable for the user.
Use Custom Service¶
Once a custom service has been successfully deployed, the user will be presented with instructions on how to access it and use it. For custom services, administrators designing the custom service need to make sure that they make this available to users. For custom services accessible via web browsers, administrators can decide to expose the service either via Ingress or a Load Balancer.
View Custom Services¶
Clicking on the custom services menu will list of all the services the user has access to, their status and additional details about them. Note that custom services may span different workspaces and different instances. To view details about a specific custom service, users just need to click on the name.
Delete Custom Service¶
To delete a custom service, users should click on the ellipses on the far right of the selected service and select delete.
Info
Once deletion has been initiated, it cannot be stopped or reversed. Users can create a new notebook if required.