Question

A CloudOps engineer manages a DynamoDB table configured with provisioned capacity. During predictable morning peaks, the application logs numerous errors and CloudWatch shows a high ThrottledRequests count while ConsumedReadCapacityUnits repeatedly hits the provisioned ceiling. During off-peak hours, consumed capacity is only 10% of provisioned. The team wants to eliminate throttling during peaks while minimizing cost the rest of the day, with the least operational overhead. What should the engineer recommend?

Accepted Answer

Enable DynamoDB Auto Scaling with a target utilization percentage so provisioned capacity adjusts to demand automatically

Answer

Manually increase the provisioned read and write capacity units to the observed daily peak and leave it fixed

Answer

Create a CloudWatch alarm on ThrottledRequests that triggers a Lambda function to reboot the application servers

Answer

Enable DynamoDB Accelerator (DAX) in front of the table to absorb all read and write traffic