Examkin

Question

Difficulty: MediumCost-Optimized Database Design and Capacity Planning

A corporate training provider is developing a new online portal. The portal requires two databases:

1. A database to cache active user session states during training modules. The traffic is highly unpredictable and experiences sudden, intense spikes when new courses are published, but is otherwise near zero.
2. A database to store employee profiles and certification histories. This workload is highly predictable and stable, with minor, gradual increases during business hours.

Which TWO database configurations should a solutions architect recommend to achieve the most cost-effective solution? (Select TWO.)

  1. Use Amazon DynamoDB in on-demand capacity mode for the session cache workload.Answer
  2. Use Amazon RDS for PostgreSQL with provisioned capacity for the relational database, and purchase database Reserved Instances for the baseline usage.Answer
  3. C
    Use Amazon DynamoDB in provisioned capacity mode for the session cache to ensure maximum throughput during sudden spikes.
  4. D
    Use Amazon RDS for PostgreSQL and configure the application to promote a read replica to the primary instance during peak write periods to handle the workload.
  5. E
    Purchase AWS Compute Savings Plans to automatically apply hourly discounts to the DynamoDB tables and Amazon RDS DB instances.

Answer

Use Amazon DynamoDB in on-demand capacity mode for the session cache, and use Amazon RDS for PostgreSQL with provisioned capacity combined with Reserved Instances for the relational database.
For the session cache, the unpredictable and spiky traffic pattern is best served by Amazon DynamoDB in on-demand mode. This ensures the system scales instantly to meet the spikes without incurring charges during idle hours. For the stable and predictable relational database, using Amazon RDS for PostgreSQL with database Reserved Instances provides the highest cost savings for a continuous baseline workload.

Step-by-Step Solution

1
Analyze the capacity requirements for the session cache workload.
The session cache workload has highly unpredictable spikes and long periods of near-zero traffic. On-demand capacity mode in DynamoDB is the most cost-effective since it avoids over-provisioning costs.
On-demand capacity mode charges only for actual read/write requests, making it cheaper than maintaining a high provisioned throughput that is rarely utilized.
2
Analyze the capacity requirements for the employee profile and certification database.
The profile database workload is relational (PostgreSQL), predictable, and continuous. Standard RDS with Reserved Instances (RIs) is the most cost-effective.
Reserved Instances provide substantial discounts for database workloads that run continuously with a predictable baseline.
3
Evaluate and rule out non-applicable discount options and architectural patterns.
Rule out Compute Savings Plans (which do not apply to RDS or DynamoDB) and manual replica promotion (which is functionally incorrect for handling write capacity and increases complexity/costs).
Ensures that the final selections align with AWS billing rules and architectural best practices.

Key Concept

Selecting optimal capacity and pricing models (On-Demand vs. Provisioned/Reserved) based on workload predictability and engine requirements.
Estimated Time:2m 0s
Rate this question

Topics

Cost-Optimized Database Design and Capacity Planning