Question

A financial services company runs a critical order-processing application backed by a DynamoDB table in us-east-1. Compliance requires the ability to recover the table to any point within the last 24 hours (RPO of 5 minutes) and to restore into a separate DR region (us-west-2) within an RTO of 2 hours in the event of a regional outage. The team wants a fully managed, low-operational-overhead solution that also protects against accidental writes or deletes. Which approach best meets these requirements?

Accepted Answer

Enable DynamoDB Point-in-Time Recovery (PITR) on the table and, when needed, use the built-in restore capability to restore the table to a new table in us-west-2 for the desired timestamp.

Answer

Configure DynamoDB global tables between us-east-1 and us-west-2 and rely on multi-region replication to recover from accidental writes and regional failures.

Answer

Schedule AWS Backup on-demand jobs every 5 minutes to back up the table and copy each recovery point to a vault in us-west-2.

Answer

Enable DynamoDB Streams and write a Lambda function that continuously exports item-level changes to an S3 bucket in us-west-2 for manual replay during recovery.