Multi-Cloud Replication: From Chaos to Control

on YouTube ↗38m

" summary

Shows how the Akka 3 SDK uses event sourcing plus append-only journals and a single-writer principle per entity to replicate state across regions and clouds, streaming discrete events over a gRPC stream pipe between clusters. Claims per-entity ownership with a handshake-based transfer so reads can be local and writes move ownership, guarantees ordering so handovers never lose pending events, and demonstrates cross-cloud replication between AWS and GCP for lower latency, higher availability and simpler disaster recovery.

AI-written summary · about

" chapters (11)

1. 0:00Introduction to multi-cloud replication
2. 3:59Why multi-cloud replication matters
3. 7:34Limitations of CRUD applications
4. 12:37Event sourcing fundamentals
5. 18:36Actor model explained
6. 24:56Location transparency benefits
7. 27:45Multi-cloud replication setup
8. 31:41Primary-region and read-only commands
9. 34:52Ownership handover process
10. 36:20Why event sourcing enables replication
11. 37:25Q&A and resources

" related talks

• 24mEngineering Real-Time Data Synchronization Across Multi-Cloud SystemsEunice Kamau · Commit Your Code 2026
• 53mDistributed Eventsourcing with AkkaDuncan K. DeVore · Philly ETE 2015
• 41mAn Adventure in Multi-Cloud with KubernetesLaurent Gil, Augustinas Stirbis · Build Stuff 2020
• 30mCONSCIOUS CONSISTENCY WITH AKKA CLUSTER, CRDTS AND DISTRIBUTED DATAJan Pustelnik · Scalar Conference 2017
• 25mCloudstate—Towards Stateful ServerlessJonas Bonér, James Roper · Ray Summit 2020
• 30mData Consistency Patterns in Cloud Native ApplicationsRyan Knight · Scale By The Bay 2018
• 52mArchitectural Patterns for Distributed and Decentralized DataJames Ward · Jfokus 2022
• 18mLeveraging Multiple Regions to Improve Site Reliability:Andrew Duch · SREcon18 Americas