Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Developer practice bank has an additional 608 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A backend service uses the Compute Engine REST API to synchronize inventory for 12,000 VM instances and update metadata across hundreds of instances. Currently, the service encounters the following issues:
aggregatedList responses stop at 500 instances.aggregatedList calls using server-side filters frequently return 403 rateLimitExceeded errors.Which combination of API optimizations should you implement to resolve these issues?
This approach combines four complementary Google Cloud API optimization techniques—token-based pagination, field masks, client-side filtering, and request batching over persistent HTTP/2 connections—to maximize data transfer efficiency while protecting API rate quotas.
maxResults. Reading nextPageToken from each response and passing it as pageToken in subsequent calls retrieves all 12,000 instances reliably.fields query parameter (such as fields=items/*/instances(name,status),nextPageToken) instructs the server to return only the requested attributes, shrinking JSON payload sizes and speeding up network transfer.*.list or *.aggregatedList methods trigger filtered_list_cost_overhead quota charges when scanning more than 10,000 resources, leading to 403 rateLimitExceeded errors. Filtering retrieved items in application memory eliminates this server-side overhead.filtered_list_cost_overhead quota exhaustion.Every symptom in the scenario is addressed at its root cause without introducing hidden quota penalties or stale data, delivering low-latency, resilient synchronization at scale.
This option attempts to bypass pagination by raising the maxResults parameter, relies on server-side filtering to trim list output, and assumes HTTP batching collapses multiple write operations into a single API quota unit.
maxResults parameter in Compute Engine list APIs has an enforced upper limit of 500 items per page. Passing 12,000 does not return the full dataset in one call; token-based iteration via pageToken is still required.filtered_list_cost_overhead rate limit when more than 10,000 resources are scanned, perpetuating 403 rateLimitExceeded errors.fields).maxResults cannot exceed 500 for Compute Engine list calls.The total inventory were under 500 instances, the API used a custom bulk endpoint billed as one unit, and no filtered_list_cost_overhead threshold existed.
A common misconception is that batching saves API quota; in reality, batching optimizes network round-trips and TCP overhead, while quota is still metered per sub-request.
This option pulls complete JSON resource representations without attribute filtering, issues high-concurrency burst traffic for updates, and uses short polling with immediate retries when errors occur.
429 or 403 rateLimitExceeded failures across the project.nextPageToken and pageToken, the inventory sync still stops after the first page of 500 results.The application required every attribute of every resource, operated well below quota thresholds, and used exponential backoff instead of short polling.
To optimize data transfer and protect quotas, developers should pair concise field masks with batching, client-side rate limiting, and truncated exponential backoff with jitter.
This option relies on long-lived local caching, assumes omitting pagination parameters triggers automatic server-side streaming of all records, and establishes a fresh TCP connection per metadata update.
pageToken) does not stream all 12,000 instances; instead, the Compute Engine API returns only the first page of up to 500 elements, leaving over 95% of the inventory unsynchronized.aggregatedList results for 24 hours prevents the synchronization service from observing real-time VM status transitions (such as RUNNING to TERMINATED).The workload consumed a static lookup dataset under 500 items that rarely changed, and used a persistent HTTP/2 connection pool for writes.
Using HTTP/2 or HTTP/1.1 keep-alive eliminates repeated handshake overhead, while token-based pagination is mandatory for reading complete multi-page datasets.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated