Google Professional Cloud Developer Question 442
Select 3Google Cloud PlatformYou are developing a serverless application on Google Cloud that processes large volumes of data using Cloud Functions. To improve efficiency and reduce costs, you want to batch multiple requests before sending them to a downstream API. Which of the following approaches should you consider to implement batching effectively within the constraints of Cloud Functions?
- A
Use Cloud Pub/Sub to aggregate messages and trigger a Cloud Function that processes them in batches.
- B
Implement a delay in your Cloud Function to accumulate requests in memory before sending them as a batch.
- C
Use Cloud Tasks to queue requests and batch them based on size or time limits.
- D
Write batch data to Cloud Storage and trigger another Cloud Function to process files as batches.
- E
Directly batch requests using the Cloud Function’s runtime memory and send them to the API when the memory limit is reached.
Show answer and explanation
Correct answers: A, C, D
Explanation
Batching requests in a serverless application on Google Cloud requires leveraging tools and services that align with the principles of statelessness and scalability. Cloud Pub/Sub, Cloud Tasks, and Cloud Storage are all reliable and scalable options for implementing batching, as they can handle large volumes of data efficiently and trigger downstream processing. Using runtime memory or adding delays within a Cloud Function is not suitable, as it contradicts the stateless nature of serverless functions and may lead to performance or cost issues.
- A. Correct.
Correct: Cloud Pub/Sub is an effective service for aggregating messages and triggering downstream processing in batch. It provides a reliable and scalable way to handle batching in a serverless architecture.
- B. Incorrect.
Incorrect: Implementing a delay in a Cloud Function to accumulate requests in memory is not recommended. Cloud Functions are stateless and designed for quick execution. Delays may lead to timeout issues or increased costs.
- C. Correct.
Correct: Cloud Tasks can be used to queue requests and trigger a batch processing mechanism based on custom criteria like size or time, making it a reliable option for batching.
- D. Correct.
Correct: Writing batch data to Cloud Storage and triggering subsequent processing via events is a valid and scalable approach for batching in serverless environments.
- E. Incorrect.
Incorrect: Using the Cloud Function's runtime memory for batching is not practical. Cloud Functions are stateless, and memory constraints could result in data loss or failures.