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 designing a storage architecture on Google Cloud for a large-scale data analytics and model-training pipeline running on a zonal GPU cluster. The application will read and write hundreds of terabytes of unstructured objects with the following requirements:
Which Cloud Storage configuration should you choose?
Cloud Storage Rapid Bucket is a specialized zonal bucket configuration that stores data in the Rapid storage class (RAPID). By colocating object storage in the exact same Google Cloud zone as your compute resources, Rapid Bucket eliminates cross-zone network hops to deliver sub-millisecond latency for open files and aggregate bandwidth scaling up to 15 TB/s. Creating a Rapid Bucket requires hierarchical namespace to be enabled at bucket creation, which organizes objects in a true directory structure and significantly boosts baseline request capacity.
Standard regional, dual-region, or multi-region buckets cannot match the sub-millisecond latency or 15 TB/s bandwidth of a zonal Rapid Bucket, and buckets without hierarchical namespace would throttle an immediate burst of 4,000 writes/sec and 20,000 reads/sec. Combining Rapid storage, hierarchical namespace, and even key distribution satisfies every performance, scalability, and cost constraint simultaneously.
A dual-region bucket stores data redundantly across two specific Google Cloud regions for high availability and cross-region resilience. Rapid Cache is a fully managed, SSD-backed zonal read cache that sits in front of an existing Cloud Storage bucket to accelerate read throughput (up to 2.5 TB/s) and lower read latency for read-heavy workloads without requiring application code changes.
The dataset was read-heavy with infrequent writes, already resided in a multi-region or dual-region bucket for cross-region disaster recovery, and required faster zonal read performance up to 2.5 TB/s without migrating data into a zonal Rapid Bucket.
The Standard storage class (STANDARD) in a regional location is designed for frequently accessed ("hot") data colocated with regional compute clusters. While Cloud Storage automatically scales bucket I/O capacity across multiple servers as load increases, standard buckets without hierarchical namespace begin with a fixed baseline capacity, and object names are indexed in lexicographical order.
YYYY-MM-DD-HH/) concentrates all writes into a narrow range of the lexicographical object index. Worse, the learned auto-scaled write capacity under a prefix resets at the top of every hour when the prefix changes, causing recurring latency spikes and throttling.The workload required regional cross-zone redundancy (99.99% typical availability), could tolerate millisecond-range rather than sub-millisecond latency, used randomized or hashed key prefixes, and gradually ramped traffic above the baseline thresholds over 20-minute intervals.
Nearline storage (NEARLINE) is a low-cost, highly durable storage class designed for infrequently accessed data read or modified roughly once per month or less. Cloud Storage FUSE is an open-source FUSE adapter that mounts Cloud Storage buckets as local file systems so applications can access objects using standard POSIX file system semantics without code refactoring.
The workload stored archival or monthly-accessed datasets (such as backups or long-tail media retained for at least 30 days) where occasional millisecond-level read latency and file-system convenience outweigh retrieval fees and sub-millisecond I/O performance.
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