Use InfinStor to manage access to MLflow Artifacts¶
MLflow uses cloud object stores (s3, azure blob store, etc.) for the actual storage of artifacts. These are accessed using API calls such as log_artifact, log_artifacts and download_artifacts
Open Source MLflow Artifact Access Control¶
In open source MLflow, access to the cloud object store used for storing MLflow artifacts is controlled by the underlying storage's authentication system. For example, in order to access s3, the user must have an AWS Access Key Id and Secret Access Key. This may also require all MLflow users to have an associated IAM user account
InfinStor MLflow Artifact Access Control¶
InfinStor MLflow includes an authorization system for artifacts that is tied to the InfinStor MLflow experiment and model Authorization system.
You can manage InfinSnap buckets by select InfinStor enabled buckets from the sidebar on the left panel.
Enabling InfinStor MLflow Artifact Access Control¶
InfinStor MLflow Artifact Access Control can be enabled at the time of single tenant stack installation, or using the InfinStor Administrator Dashboard
Using the InfinStor MLflow Artifact Access Control¶
End users need to import the infinstor_mlflow_plugin in their python code.