MLflow Model Serving on Azure Databricks
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-13-2022 09:01 AM
I know that in the documentation about model serving says.
- The cluster is maintained as long as serving is enabled, even if no active model version exists. To terminate the serving cluster, disable model serving for the registered model.
- The cluster is considered an all-purpose cluster, subject to all-purpose workload pricing.
- Global init scripts are not run on serving clusters.
In order to try to save cost, I want to ask if someone know or has checked a way to start/stop those Job clusters via Scheduling feature in ADB?
Can we use same cluster for different model?
Labels:
- Labels:
-
ADB
-
Azure
-
Azure databricks
-
Global init scripts