Runtime Configuration


  • A Kubeflow Pipelines Endpoint
  • IBM Cloud Object Storage or other S3 Based Object Store (Optional)

Configuring Runtime Metadata

AI Pipelines require configuring a pipeline runtime to enable its full potential. AI Pipelines currently only support Kubeflow Pipelines with plans to expand support for other runtimes in the future.

Using Command Line Interface

To configure runtime metadata for Kubeflow Pipelines use the elyra-metadata install runtimes command providing appropriate options. This command will create a json file in your local Jupyter Data directory under its metadata/runtimes subdirectories. If not known, the Jupyter Data directory can be discovered by issuing a jupyter --data-dir command on your terminal.

Here’s an example invocation of elyra-metadata install runtimes --schema_name=kfp to create runtime metadata for use by Kubeflow Pipelines corresponding to the example values in the table below. Following its invocation, a file containing the runtime metadata can be found in [JUPYTER DATA DIR]/metadata/runtimes/my_kfp.json.

elyra-metadata install runtimes --schema_name=kfp \
       --name=my_kfp \
       --display_name="My Kubeflow Pipeline Runtime" \
       --api_endpoint= \ \
       --api_password=mypassword \
       --cos_endpoint=http://minio-service.kubeflow:9000 \
       --cos_username=minio \
       --cos_password=minio123 \

This produces the following content in my_kfp.json:

    "display_name": "My Kubeflow Pipeline Runtime",
    "schema_name": "kfp",
    "metadata": {
        "api_endpoint": "",
        "api_username": "",
        "api_password": "mypassword",
        "cos_endpoint": "http://minio-service.kubeflow:9000",
        "cos_username": "minio",
        "cos_password": "minio123",
        "cos_bucket": "test-bucket"

NOTE: In case of typing a custom bucket name using minio cloud storage, make sure the bucket name has no underscores

To validate your new configuration is available, run:

elyra-metadata list runtimes

Available metadata instances for runtimes (includes invalid):

Schema   Instance  Resource  
------   --------  -------- 
kfp      my_kfp    /Users/jdoe/Library/Jupyter/metadata/runtimes/my_kfp.json

Existing runtime metadata configurations can be removed via elyra-metadata remove runtimes --name=[runtime]:

elyra-metadata remove runtimes --name=my_kfp

Elyra depends on its runtime metadata to determine how to communicate with your KubeFlow Pipelines Server and with your chosen Object Store to store artifacts.

Using Elyra Runtimes User Interface

Elyra also has UI support to accessing runtime metadata directly in JupyterLab, where user can easily view, add, edit or remove runtimes configurations, as an option to the CLI method described above. It can be accessed from the Runtimes icon on JupyterLab’s left tab bar, which opens a side panel widget displaying available pipeline runtimes. Click on the + button to add a new pipeline runtime, or on the edit icon to edit an existing configuration. These actions will open the Metadata Editor as a new tab in the main area, where runtime metadata parameters can be edited and will also be saved in json format in [JUPYTER DATA DIR]/metadata/runtimes/. Runtimes UI



The KubeFlow Pipelines API Endpoint you wish to run your Pipeline.



Username used to access your KubeFlow Pipelines API endpoint. SEE NOTE.



Password used to access your KubeFlow Pipelines API endpoint. SEE NOTE.

Example: mypassword


This should be the URL address of your S3 Object Storage. If running an Object Storage Service within a kubernetes cluster (Minio), you can use the kubernetes local DNS address.

Example: https://minio-service.kubeflow:9000


Username used to access the Object Store. SEE NOTE.

Example: minio


Description: Password used to access the Object Store. SEE NOTE.

Example: minio123


Name of the bucket you want your artifacts in. If the bucket doesn’t exist, it will be created

Example: test-bucket

NOTE: If using an authentication-restricted Kubeflow environment, you must enter your credentials in api_username and api_password fields to allow pipeline uploads through Elyra.

If using IBM Cloud Object Storage, you must generate a set of HMAC Credentials and grant that key at least Writer level privileges. Your access_key_id and secret_access_key will be used as your cos_username and cos_password respectively.