Google Professional Machine Learning Engineer Question 385
Single answerGoogle Cloud PlatformYou are developing a machine learning web application that serves predictions via a REST API. The model inference is containerized, and you need to deploy it on a fully managed, serverless platform with minimal operational overhead. The solution should automatically scale based on traffic and integrate easily with other Google Cloud services. Which Google Cloud service is the most appropriate for this use case?
- A
Cloud Run
- B
Compute Engine
- C
Kubernetes Engine
- D
App Engine
Show answer and explanation
Correct answer: A
Explanation
Cloud Run is the best choice for deploying a containerized machine learning model as a REST API in a serverless manner. It is fully managed, scales automatically based on traffic, and integrates seamlessly with other Google Cloud services. This makes it ideal for reducing operational overhead, especially for machine learning workloads where traffic can vary significantly.
- A. Correct.
Cloud Run is the correct choice as it is a fully managed, serverless platform that can run containerized applications. It automatically scales based on traffic, is cost-efficient, and integrates well with other Google Cloud services.
- B. Incorrect.
Compute Engine is not a suitable choice for this use case because it requires managing virtual machines, which increases operational overhead compared to serverless options.
- C. Incorrect.
Kubernetes Engine provides container orchestration but requires managing clusters and nodes, which is more complex than the serverless approach of Cloud Run.
- D. Incorrect.
App Engine is designed for applications written in specific runtime environments and may not support the flexibility of running containerized applications as effectively as Cloud Run.