Post migration tasks with CDSW migration tool and command-line utility tool

The post-migration tasks apply to both migration tools and have to be considered after the migration.

Perform the following post-migration actions:

• Validate that your code can run on Cloudera AI, especially if the projects are switching from Legacy Engines to ML Runtime images. Differences exist between the built-in Python packages in the Legacy image compared to the Runtime image.
• All Models, Jobs, and Applications in the migrated projects will be in a suspended or stopped state in the target Cloudera AI Workbench, therefore all the artifacts must be restarted post migration. Before starting the Models, Jobs, or Applications in the destination workbench, the corresponding workloads must be stopped in the source workbench to avoid any data corruption if both workbenches are accessing the same data.
• Verify and update, if required, the ML Runtime image and the Spark Addon for each Model, Job, and Application in the migrated projects.
• Migrated projects do not have Spark Pushdown feature enabled. If required, the Spark Pushdown feature must be enabled at project level. Currently, this cannot be set globally for all projects but changing this for all projects in the Cloudera AI database is possible.
• Cloudera AI Applications' URL will be different. Users of the applications must be informed of the new URLs.