Exam Professional Cloud Database Engineer topic 1 question 22 discussion

Correct answer is D. https://cloud.google.com/sql/docs/mysql/best-practices#data-arch - Split your large instances into smaller instances, where possible.

The solution D is better if we can execute a massive refactor. So, the best solution is C because if CPU is normal, the issue is the capacity of the I/O. For increase it, it's enough increase the disk storage. Source: https://cloud.google.com/sql/docs/sqlserver/best-practices

1. Microservices Best Practices: • The database-per-service pattern reduces contention by isolating workloads. Each microservice uses a dedicated Cloud SQL instance, preventing cross-service resource competition. • Smaller instances are easier to scale independently based on individual service needs. 2. Horizontal Scaling: • Splitting the monolithic database into smaller instances aligns with Google’s recommendation for horizontal scaling in microservices (search result). • Avoids hitting connection limits, I/O bottlenecks, or lock contention that may occur in a single large instance, even if CPU utilization appears normal. 3. Minimal Refactoring: • Requires only configuration changes (e.g., updating connection strings) rather than rewriting application logic.

Storage Size Impact: Increasing the storage size for your Cloud SQL instance can improve performance because Cloud SQL automatically increases the IOPS (Input/Output Operations Per Second) as the storage size increases. This can help resolve performance bottlenecks related to disk I/O, which might be the underlying issue given that CPU utilization is normal. No Major Refactoring: This approach does not require any significant changes to your application architecture or code, making it a straightforward solution. Scalability: By increasing the storage size, you can ensure that your instance has the necessary IOPS to handle the increasing load as your application grows.

cant do major refactoring. also CPU utilization is under control then mostly issue is with IOPS capacity of CloudSQL we need to increase storage space. to increase IOPS

d

Agree with C as the CPU Looks normal.

Splitting smaller databases require major effort. Answer should be C

It's a microservices architecture and CPU utilization is normal. This means that having multiple Cloud SQL instances will help for each microservice.

Wouldn't D be a big refactoring, as well as switching to Spanner, especially if MySQL is considered? This question is bad.

D. Needing to avoid any major refactoring eliminates A. The question states CPU is not an issue, so that eliminates B. Adding more storage would increase IOPS, but there’s no indication network throughput is an issue, so that eliminates C. That leaves D. A microservice architecture is supposed to use a separate database for each microservice, rather than one big database for all the microservices. So D it is. The link provided by Kloudgeek is spot on.

D: Use many smaller Cloud SQL instances.

D: Use many smaller Cloud SQL instances.

D: Use many smaller ***** Cloud SQL instances.

D is the correct answer

Split CloudSql instance into many small instances to support Microservices