Exam Professional Cloud Database Engineer topic 1 question 54 discussion

I'd choose B. To me, there is no totally true option. I agree with dynamic_dba that A and D are wrong. B. Spanner supports JSON, but not that good (as Pilot50 pointed out) C. Firestore Firestore in Datastore mode does provide some degree of consistency guarantees, specifically "eventual consistency" for global queries and strong consistency for ancestor queries. However, it does not provide strong consistency for non-ancestor queries. In contrast, Cloud Spanner provides strong consistency for all transactions, which is especially important for applications that require ACID transactions across multiple regions. So between B and C, I'll pick Spanner for strong consistency

I choose option C. Spanner could work, but it would be very tight. Firestore is the best fit for the use case since it auto-scales transparently and works perfectly with semi-structured data

b: https://cloud.google.com/spanner/docs/working-with-json#spanner_query_with_json_parameter-java

A. Use Cloud SQL for PostgreSQL with both cross-region read replicas: This won't scale transparently or globally B. Use Cloud Spanner in a multi-region configuration: supports JSON, designed for mission critical workloads, can be scaled transparently with an autoscaler across the globe. C. Use Firestore in Datastore mode: Firestore can scale transparently in a multi-region; but it can't be scaled GLOBALLY. D. Use a Bigtable instance with clusters in multiple regions: This doesn't necesarily fit the supplied workload.

Spanner can't support JSON and not scaling automatically

C. Cloud SQL is a regional service so cannot scale to multiple regions. Eliminate A. Spanner doesn’t support semi-structured data and doesn’t scale transparently. Eliminate B. Bigtable doesn’t work with JSON data. Eliminate D. That leaves C, Firestore in Datastore mode which satisfies all the requirements.

Hello Team, I hope this comment helps Work with transactions and offer strong consistency Work with structured and semi-structured (JSON) data => NoSQL (cloud SQL + spanner are not suitable here ) Scale transparently to multiple regions globally as the operation grows => scale transparently => you don't manage anything Big Table is for time series, millisecond access and for JSON Referrence ; https://cloud.google.com/firestore#all-features

Why not bigtable ?

Spanner is multiple region and can Scale transparently

A and B are correct answer but B looks best answer because of "Scale transparently to multiple regions globally as the operation grows."

A: Use Cloud SQL for PostgreSQL ***** with both cross-region read replicas.

I'll go for Spanner because of 'Scale transparently to multiple regions globally as the operation grows'