Examkin

Question

Difficulty: Very hardSelecting and Designing Storage and Database Resources

Match each application workload access pattern and architectural requirement to the most appropriate Google Cloud storage or database service.

  • High-throughput time-series metric ingestion exceeding 1,000,0001,000,000 writes per second requiring low single-digit millisecond latency key-value reads by timestamp.Cloud Bigtable
  • Globally distributed relational OLTP database requiring full ANSI SQL support, multi-region ACID transactions, and a 99.999%99.999\% availability SLA.Cloud Spanner
  • Sub-millisecond latency in-memory data store supporting complex data structures, geospatial queries, and session caching.Memorystore for Redis
  • Serverless document store offering automatic offline client sync, live query snapshot listeners, and multi-region atomic transactions for web and mobile frontends.Firestore

Answer

High-throughput time-series metric ingestion matches Cloud Bigtable. Globally distributed relational OLTP with multi-region ACID guarantees matches Cloud Spanner. Sub-millisecond in-memory caching with geospatial support matches Memorystore for Redis. Serverless document storage with offline sync matches Firestore.
Each workload specification aligns directly with the architectural design principles of GCP database offerings: Cloud Bigtable for high-scale write-heavy time-series, Cloud Spanner for globally distributed relational ACID transactions, Memorystore for Redis for sub-millisecond in-memory operations, and Firestore for serverless real-time document synchronization.

Step-by-Step Solution

1
Analyze high-throughput time-series ingestion requirement (>1,000,000 writes/sec)
Identify Cloud Bigtable as the optimal scalable NoSQL wide-column store designed for time-series metrics and heavy write throughput.
Cloud Bigtable scales linearly to handle millions of IOPS with consistent low-latency key lookups.
2
Evaluate global relational OLTP requirement with a 99.999% availability SLA
Identify Cloud Spanner as the target managed relational database.
Cloud Spanner uses TrueTime hardware clocks and Paxos consensus to deliver multi-region strong consistency and five-nines availability.
3
Examine sub-millisecond caching requirement with geospatial query capabilities
Identify Memorystore for Redis.
Memorystore for Redis delivers ultra-low latency in-memory data management supporting advanced Redis primitives like GEO indexing.
4
Review mobile/web frontend document store requirement with offline sync and live listeners
Identify Firestore.
Firestore client SDKs natively support real-time data sync, offline data caching, and automatic scaling for app developers.

Key Concept

Evaluating workload access patterns, consistency requirements, scale, and latency targets to select optimal GCP storage and database resources.
Rate this question

Topics

Selecting and Designing Storage and Database Resources