SAA-C03 Question 373
Single answerA startup is hosting a serverless web application on AWS and wants to optimize costs while maintaining performance. The application uses Amazon API Gateway, AWS Lambda functions, and Amazon DynamoDB. The DynamoDB table's write throughput is highly variable, with periods of intense activity followed by long periods of low or no writes. What is the most cost-effective way to configure the DynamoDB table to handle this workload?
- A
Configure the DynamoDB table to use provisioned capacity with sufficient write capacity units to handle peak traffic.
- B
Enable DynamoDB Auto Scaling on the table to dynamically adjust the provisioned capacity based on traffic patterns.
- C
Switch the DynamoDB table to on-demand capacity mode.
- D
Use global secondary indexes (GSIs) to reduce the write traffic on the base table.
Show answer and explanation
Correct answer: C
Explanation
For workloads with unpredictable and variable traffic patterns, DynamoDB's on-demand capacity mode is the most cost-effective solution. It eliminates the need to provision capacity and charges only for actual read and write requests. This is ideal for the startup's scenario where write activity fluctuates significantly.
- A. Incorrect.
Configuring the table with provisioned capacity for peak traffic leads to paying for unused capacity during periods of low activity, which is not cost-effective.
- B. Incorrect.
While Auto Scaling can dynamically adjust capacity, it is still tied to provisioned capacity. During periods of no traffic, there could still be costs for maintaining the minimum provisioned capacity.
- C. Correct.
On-demand capacity mode is cost-effective for workloads with unpredictable and sporadic traffic patterns, as it charges only for the read and write requests performed without provisioning capacity.
- D. Incorrect.
Global secondary indexes (GSIs) improve query performance but do not directly reduce the cost of handling variable write throughput. They may even add additional costs for maintaining the index.