nerdexam
Google

PROFESSIONAL-CLOUD-DEVELOPER · Question #326

You are developing a new API that creates requests on an asynchronous message service. Requests will be consumed by different services. You need to expose the API by using a gRPC interface while minim

The correct answer is B. Deploy your API as a Cloud Run service. Create a Pub/Sub topic, and configure your API to push. The two requirements are: gRPC interface support and minimal infrastructure management overhead. Cloud Run is a fully managed serverless platform that natively supports gRPC (HTTP/2), requires zero cluster or VM management, and scales to zero. Pub/Sub is Google's fully managed as

Developing and Deploying Applications

Question

You are developing a new API that creates requests on an asynchronous message service. Requests will be consumed by different services. You need to expose the API by using a gRPC interface while minimizing infrastructure management overhead. How should you deploy the API?

Options

  • ADeploy your API to App Engine. Create a Pub/Sub topic, and configure your API to push
  • BDeploy your API as a Cloud Run service. Create a Pub/Sub topic, and configure your API to push
  • CDeploy your API to a GKE cluster. Create a Kafka cluster, and configure your API to write
  • DDeploy your API on a Compute Engine instance. Create a Kafka cluster, and configure your API

How the community answered

(54 responses)
  • A
    4% (2)
  • B
    72% (39)
  • C
    7% (4)
  • D
    17% (9)

Explanation

The two requirements are: gRPC interface support and minimal infrastructure management overhead. Cloud Run is a fully managed serverless platform that natively supports gRPC (HTTP/2), requires zero cluster or VM management, and scales to zero. Pub/Sub is Google's fully managed asynchronous messaging service, eliminating the need to manage a broker like Kafka. Option A (App Engine) has less flexible gRPC support and higher management overhead than Cloud Run. Options C and D require managing Kafka clusters (significant operational overhead) on top of GKE or Compute Engine respectively - the opposite of minimal management.

Topics

#Serverless Deployment#gRPC APIs#Asynchronous Messaging#Managed Services

Community Discussion

No community discussion yet for this question.

Full PROFESSIONAL-CLOUD-DEVELOPER Practice