Amazon S3 Replication: For data protection & application acceleration
summary
Amazon S3 replication automatically copies objects between buckets, preserves original metadata and version IDs, and can replicate to multiple destinations (same or cross-region) including direct writes to Glacier/Deep Archive. Replication Time Control replicates most objects in seconds and 99.99% within 15 minutes and is backed by an SLA; it emits events and CloudWatch metrics for failures and latency. Cross-account replicas can change ownership, require versioning plus an IAM role grant, and support delete marker replication and replica modification sync for bi‑directional active‑active setups.
AI-written summary · about
chapters (9)
- 0:00Introduction to S3 Replication
- 2:31Why use S3 replication
- 4:21Replication Time Control
- 7:07Multi-destination replication
- 7:49Setting up replication
- 10:37Advanced replication options
- 12:43Active-active replication
- 13:03Customer use cases
- 15:15Summary and Q&A
related talks
48mReplication as a First-Class Citizen: The Art of Global Resilience
38mMulti-Cloud Replication: From Chaos to Control
18mLeveraging Multiple Regions to Improve Site Reliability:
52mHow We Survive Split-Brain and Keep Writing Data via the S3 Protocol
22mReliable Data Replication
01h 08mTransactions and Scalability in Cloud Databases—Can’t We Have Both?
24mDistributed Transactions at Scale in Amazon DynamoDB
35mKeeping Ceph RGW Object Storage Consistent