What underlying mechanisms govern AWS S3 Cross-Region Replication when engineering low-latency data access and automated multi-region disaster recovery systems? Furthermore, why is configuring asynchronous, policy-driven object synchronization across geographically separated AWS Regions the definitive approach to achieving high data durability and strict regulatory compliance?