Answer-first summary for fast verification
Answer: Expose each individual model as an endpoint in Vertex AI Endpoints. Use Cloud Run to orchestrate the workflow.
Option C is the best choice because it leverages Vertex AI Endpoints for version control of each individual model and uses Cloud Run for orchestrating the workflow. This combination ensures easy scaling to zero, minimizing compute resource utilization and automating much of the management process. Other options either add unnecessary complexity or utilize tools (like BigQuery) that are not optimized for model version control.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have developed an application for your company's products that utilizes a sequence of scikit-learn models to predict optimal pricing. The workflow consists of multiple steps, each handled by individual models that are also used in other workflows by team members. You aim to deploy this workflow with robust version control for each model as well as the entire workflow. The application needs to be scalable, with the ability to scale down to zero when not in use, to optimize compute resource utilization and reduce manual management efforts. What approach should you take?
A
Expose each individual model as an endpoint in Vertex AI Endpoints. Create a custom container endpoint to orchestrate the workflow.
B
Create a custom container endpoint for the workflow that loads each model’s individual files. Track the versions of each individual model in BigQuery.
C
Expose each individual model as an endpoint in Vertex AI Endpoints. Use Cloud Run to orchestrate the workflow.
D
Load each model’s individual files into Cloud Run. Use Cloud Run to orchestrate the workflow. Track the versions of each individual model in BigQuery.
No comments yet.