Associate Cloud Engineer ยท Choosing and deploying data services
Selecting a database service
Choosing among Cloud SQL, AlloyDB, Spanner, Firestore, Bigtable, BigQuery and Memorystore based on data model, scale, consistency and latency needs.
Key points
- Cloud SQL: standard relational, managed MySQL/PostgreSQL/SQL Server.
- AlloyDB: PostgreSQL-compatible, faster, transactional/analytical mix.
- Spanner: relational, horizontal scale, strong consistency, multi-region option.
- Firestore: serverless document store, app backends.
- Bigtable: wide-column, huge throughput, low latency, no joins.
- Memorystore: in-memory cache.
- BigQuery: serverless SQL analytics warehouse, not transactional.
Source: https://cloud.google.com/products/databases
Practice questions
Which service is a serverless data warehouse meant for large-scale SQL analytics?
- Cloud SQL
- Bigtable
- BigQuery
- Memorystore
Show answer
Answer: C. BigQuery
BigQuery is the serverless warehouse. Cloud SQL and Spanner serve transactional workloads, and Bigtable is a low-latency NoSQL store.
Which of these services are relational databases that speak SQL?
- Cloud SQL
- Bigtable
- AlloyDB
- Memorystore
- Spanner
Show answer
Answer: A. Cloud SQL, C. AlloyDB, E. Spanner
Cloud SQL, AlloyDB and Spanner are relational. Bigtable is wide-column NoSQL and Memorystore is an in-memory cache.
True or false: Bigtable is a good fit for relational workloads that need multi-table joins.
Show answer
Answer: False
False. Bigtable is a wide-column NoSQL store built for huge key-based throughput and low latency, not for joins.
Practice 8 questions on this topic in Bitzu
Short lessons, instant explanations and spaced review of what you get wrong. Free on iOS and Android.
