You are designing a data processing system that requires data replication across multiple AWS regions. Which AWS storage service should you use, and how should you configure it to optimize for cross-region replication?