
Ultimate access to all questions.
As the lead ML engineer at a retail company, your team is tasked with implementing a centralized system to track and manage ML metadata for generating reproducible experiments and artifacts. The system must support scalability for hundreds of concurrent experiments, ensure data lineage for compliance, and integrate seamlessly with existing Google Cloud services. Which solution would you recommend? (Choose one correct option)
A
Utilize the Hive Metastore for storing relational entities, leveraging its compatibility with Hadoop ecosystems.
B
Implement Google Cloud‘s Operations Suite for ML metadata storage, taking advantage of its logging and monitoring capabilities.
C
Adopt Vertex ML Metadata for managing ML workflows, benefiting from its comprehensive tools for tracking, managing, and storing ML metadata and artifacts.
D
Store tf.logging data in BigQuery, utilizing its analytics capabilities for ML experiment tracking.