A cloud architecture team at a healthcare organization is evaluating Google Cloud serverless options for two distinct microservices:
1. Service A: A legacy enterprise Java application packaged as a custom Docker container image listening on port 8080. It requires request concurrency (handling multiple concurrent HTTP requests per instance) and high memory allocation (up to 16 GB RAM).
2. Service B: A lightweight event-driven background handler written in Node.js that runs for under 30 seconds to extract metadata whenever a new patient record file is uploaded to a Cloud Storage bucket.
Which TWO deployment decisions correctly align with Google Cloud serverless compute best practices for these requirements? (Select TWO.)
- Deploy Service A to Cloud Run because it supports custom container images, multi-request concurrency per instance, and high memory configurations.Answer
- BDeploy Service A to Cloud Functions (1st gen) because it automatically converts legacy custom container images into serverless function packages with multi-request concurrency.
- Deploy Service B to Cloud Functions (or Cloud Functions 2nd gen) triggered by Cloud Storage bucket events to execute code directly without maintaining container infrastructure.Answer
- DDeploy Service B to Compute Engine Managed Instance Groups with an External HTTP Load Balancer to capture Cloud Storage file upload events natively.
Answer
The correct architectural decisions are deploying Service A to Cloud Run and deploying Service B to Cloud Functions.
Deploying Service A to Cloud Run is correct because Cloud Run fully supports custom Docker containers, configurable memory up to 32 GB, and request concurrency on a single instance. Deploying Service B to Cloud Functions is correct because it natively integrates with Cloud Storage events to execute lightweight event handlers without needing dedicated container builds or compute management.
Step-by-Step Solution
Key Concept
Selecting serverless compute execution paradigms based on container requirements, concurrency settings, and event trigger mechanisms.
Estimated Time:2m 0s