AWS Certified Database – Specialty — Question 290

A company has an application that uses an Amazon DynamoDB table as its data store. During normal business days, the throughput requirements from the application are uniform and consist of 5 standard write calls per second to the DynamoDB table. Each write call has 2 KB of data.
For 1 hour each day, the company runs an additional automated job on the DynamoDB table that makes 20 write requests per second. No other application writes to the DynamoDB table. The DynamoDB table does not have to meet any additional capacity requirements.
How should a database specialist configure the DynamoDB table's capacity to meet these requirements MOST cost-effectively?

Answer options

• A. Use DynamoDB provisioned capacity with 5 WCUs and auto scaling.
• B. Use DynamoDB provisioned capacity with 5 WCUs and a write-through cache that DynamoDB Accelerator (DAX) provides.
• C. Use DynamoDB provisioned capacity with 10 WCUs and auto scaling.
• D. Use DynamoDB provisioned capacity with 10 WCUs and no auto scaling.

Correct answer: C

Explanation

Each standard write of 2 KB requires 2 WCUs, meaning the baseline load of 5 writes per second requires 10 WCUs (5 * 2). During the daily 1-hour automated job, the throughput demand increases by an additional 40 WCUs (20 writes/second * 2 WCUs), making auto scaling with a baseline of 10 WCUs the most cost-effective choice to dynamically scale up capacity and avoid throttling. Starting with 5 WCUs is insufficient even for baseline traffic, and DynamoDB Accelerator (DAX) is primarily a read cache that does not optimize write capacity.