Google Professional Cloud Database Engineer Question 164
Select 2Google Cloud PlatformYour team manages a Cloud Spanner database for a global e-commerce platform. During peak shopping seasons, the demand increases significantly, requiring the database to handle more queries and accommodate larger datasets. What should you do to ensure the database scales effectively while maintaining performance?
- A
Add more nodes to the Cloud Spanner instance.
- B
Increase the CPU and memory of Cloud Spanner nodes.
- C
Partition the data using additional columns in primary keys to improve distribution.
- D
Switch to a regional configuration to optimize costs.
- E
Enable and configure horizontal autoscaling for Cloud Spanner.
Show answer and explanation
Correct answers: A, C
Explanation
To scale a Cloud Spanner database effectively during peak loads, you need to scale out by adding more nodes. Additionally, optimizing data distribution by partitioning the data using additional columns in the primary key ensures that data is evenly distributed across the nodes, improving query performance and scalability. Cloud Spanner's scaling model is based on horizontal scaling rather than vertical scaling or autoscaling.
- A. Correct.
Adding more nodes to the Cloud Spanner instance increases the compute and storage capacity, allowing Cloud Spanner to handle more queries and larger datasets. This is a key approach for scaling out.
- B. Incorrect.
Cloud Spanner does not allow manual adjustment of CPU and memory for its nodes; it is a managed service where scaling is achieved by adding nodes or optimizing data distribution.
- C. Correct.
Partitioning the data using additional columns in primary keys improves data distribution across nodes, which helps in scaling out horizontally and maintaining performance.
- D. Incorrect.
Switching to a regional configuration reduces costs but sacrifices high availability and global distribution, which are critical for a global e-commerce platform.
- E. Incorrect.
Cloud Spanner does not have a native horizontal autoscaling feature. Scaling is managed by manually adding or removing nodes.