PROFESSIONAL-CLOUD-DATABASE-ENGINEER · Question #3
Your application follows a microservices architecture and uses a single large Cloud SQL instance, which is starting to have performance issues as your application grows. in the Cloud Monitoring…
The correct answer is D. Use many smaller Cloud SQL instances. Using many smaller Cloud SQL instances is the Google-recommended practice for microservices architectures. The database-per-service pattern aligns each microservice with its own dedicated Cloud SQL instance, preventing one service's workload from starving others and eliminating…
Question
Your application follows a microservices architecture and uses a single large Cloud SQL instance, which is starting to have performance issues as your application grows. in the Cloud Monitoring dashboard, the CPU utilization looks normal You want to follow Google-recommended practices to resolve and prevent these performance issues while avoiding any major refactoring. What should you do?
Options
- AUse Cloud Spanner instead of Cloud SQL.
- BIncrease the number of CPUs for your instance.
- CIncrease the storage size for the instance.
- DUse many smaller Cloud SQL instances.
How the community answered
(37 responses)- A11% (4)
- B5% (2)
- C3% (1)
- D81% (30)
Explanation
Using many smaller Cloud SQL instances is the Google-recommended practice for microservices architectures. The database-per-service pattern aligns each microservice with its own dedicated Cloud SQL instance, preventing one service's workload from starving others and eliminating connection contention on a single large instance. Since CPU utilization is normal, the bottleneck is likely connection limits, lock contention, or I/O saturation - not raw compute. Adding CPUs (B) would not help since CPU is not the bottleneck. Increasing storage (C) does not resolve performance issues. Migrating to Cloud Spanner (A) would constitute major refactoring, which the question explicitly rules out.
Topics
Community Discussion
No community discussion yet for this question.