For migrating legacy on-prem storage with large data volumes, which pattern helps minimize downtime?

• A. Use AWS DMS for ongoing replication for initial load, with a cutover plan; optionally Snowball/Storage Gateway for initial data transfer.
• B. Migrate all data and services in a single weekend without replication.
• C. Use only network optimization to speed transfers.
• D. Use manual data validation after cutover.

Answer: (A)

Minimizing downtime during a large on-prem to AWS migration hinges on keeping the source and target in sync until the moment you cut over. The best pattern uses ongoing replication to load the initial data while the on-prem system remains active, plus a well-planned cutover to redirect traffic to the cloud. AWS DMS supports change data capture, so it can continuously replicate changes from on-prem to the target during the migration. This means the destination stays up to date, and when you flip the switch, only a small delta needs to be applied, keeping the cutover brief. For very large datasets, you can transfer the initial bulk offline—using Snowball or Storage Gateway—to reduce network impact, then resume replication to catch any remaining changes until cutover.

A plan that tries to move everything in one weekend without replication would require long downtime and risks data drift. Relying only on network speed won’t guarantee synchronization, and performing manual data validation after cutover tends to extend the downtime rather than shorten it.