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.
You are architecting a globally distributed media application that uses two storage services across multiple regions:
Immediately after an application writes new session data to Bigtable and uploads a new version of an object to Cloud Storage from the same region, a regional outage strikes that source region before cross-region replication finishes.
How do Bigtable and Cloud Storage behave when clients attempt to read the recently written data from another region?
This option reverses the actual consistency behaviors of the two services, claiming that Bigtable blocks reads to prevent stale data while Cloud Storage serves stale object versions to maintain availability.
500 error until the source region recovers.This behavior would only occur if Bigtable implemented synchronous quorum locks across regions (which it does not) and Cloud Storage used eventually consistent object versioning.
Understanding CAP theorem tradeoffs in Google Cloud is essential: multi-cluster Bigtable prioritizes availability and low latency (eventual consistency), whereas Cloud Storage guarantees strong global consistency for objects and metadata, returning a transient 500 error when unreplicated bytes are unreachable.
This option claims that having one Bigtable cluster per region guarantees global read-your-writes consistency even during failover, and that Cloud Storage exhibits eventual consistency for object metadata and listings, resulting in 404 Not Found errors.
404 Not Found for that object, even if cross-region payload replication has not finished.500 error—not a 404—to indicate temporary unavailability rather than non-existence.The Bigtable statement would hold true only while the source region remained healthy and no failover happened, and the 404 behavior would apply only to legacy eventually consistent object stores or objects deleted before a read.
Always distinguish between publicly cached objects (controlled by Cache-Control headers, which can serve stale cached content) and direct Cloud Storage operations, which are strongly consistent.
This option asserts that Bigtable consistency tokens automatically preserve read-your-writes consistency during multi-cluster failover, and that Cloud Storage performs synchronous cross-region replication for both object payloads and metadata.
CheckConsistency) is an explicit API mechanism used by applications to poll and verify whether all writes made prior to creating the token have finished replicating across clusters. It does not route read traffic or enforce synchronous failover consistency. Once failover redirects clients to another region, read-your-writes consistency is lost for unreplicated writes.CheckConsistency immediately after creating a token can take up to a few minutes while clusters verify replication status.This option would be closer to accurate only if both services used cross-region synchronous data replication (such as Spanner for relational data), which pays higher write latency in exchange for zero RPO and global strong consistency.
To reduce the asynchronous replication window in Cloud Storage, developers can use dual-region buckets with turbo replication, which targets a 15-minute RPO for object data, though replication remains asynchronous.
Cloud Bigtable uses an asynchronous, eventually consistent replication model across clusters in a replicated instance. Cloud Storage, by contrast, enforces strong global consistency for object uploads, overwrites, deletions, and metadata operations across all bucket location types, including multi-region buckets.
404 Not Found and instead returns a retryable 500 error.This choice accurately contrasts how Bigtable trades consistency for availability during cross-region failover with how Cloud Storage preserves strong consistency by returning a retryable 500 error when an unreplicated payload is temporarily unreachable.
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