GCP Professional Cloud Architect Practice Question

Your retail company runs a containerized payment-notifications service that sits idle most of the day, but traffic spikes from zero to roughly 2,000 HTTP POST requests per second for a few minutes whenever flash-sales end. Each request completes in under 300 ms, and you want to keep 95th-percentile latency under 500 ms while paying nothing when the service is idle. You do not want to manage any servers or clusters. Which Google Cloud offering best meets these autoscaling and operational requirements?

App Engine flexible environment with automatic scaling enabled.
A Google Kubernetes Engine deployment using the Horizontal Pod Autoscaler.
Cloud Run with its default automatic request-based autoscaling.
A Compute Engine managed instance group that scales VMs based on CPU utilization.

Report Issue

Answer Description

Cloud Run automatically scales container instances from zero up to the number needed to meet per-request concurrency limits, so it can absorb sudden bursts of thousands of requests per second while charging only for the CPU, memory, and request time consumed. Compute Engine managed instance groups and GKE deployments can also be configured to scale to zero compute capacity, but they still require you to manage VM images, operating-system patching, and scaling policies (and GKE adds cluster administration). App Engine flexible environment always keeps at least one instance running, incurring standby cost. Because Cloud Run eliminates all infrastructure management while providing request-driven scaling and a pay-per-use model, it best satisfies the stated requirements.

Ask Bash

Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.