You work at BrightAd Media , an advertising agency where you've built a Spark ML model to predict click-through rates for ad placements. The entire solution has been developed in your on-premises data center, but the company is rapidly migrating to Google Cloud as the data center is being decommissioned. As part of the migration: You need to continue retraining the Spark ML models. The training data is being moved to BigQuery . A fast lift-and-shift migration is required without fully rearchitecting the pipeline. How should you move your training workflow to Google Cloud?