Event sourcing 3
Download as PPTX, PDF
0 likes246 views
Event sourcing is an architecture pattern where all changes to application state are stored as a sequence of events. These events can be used to reconstruct past states and serve as the system of record. The key aspects of event sourcing are storing all changes as a sequence of immutable events, maintaining a write-optimized event log, and deriving the current state by replaying the events in order. Event sourcing allows for easy duplication of data through event replay and makes it possible to undo changes by replaying a modified sequence of events. Popular technologies for implementing event sourcing include Event Store and Apache Kafka.
Event sourcing 3
- 2. Stats about me Name: Radu Iscu Years of copy-pasting code professionally: 10 Technologies outlived: .net 1.x, Flash, Flex, Web Forms, Web parts Programming languages outlived: Turbo Pascal, Perl, PHP, Action Script 2-3, Visual Basic .net Programming languages: c#, javascript, scala
- 3. What is event sourcing?
- 5. Often talked about, never explained
- 7. Event store
- 8. WORM log of events that happened
- 9. Examples
- 10. Bank account
- 11. Patient journal
- 12. Patient journal
- 14. Multiple kinds of projections
- 17. It is replayable and undoable
- 18. It is easily duplicated
- 19. Easy to duplicate through replay
- 20. Index Events by user ID Events by company ID Events by event type Events by event sender
- 21. Snapshots
- 22. Hydrator
- 23. Cache
- 24. Event versioning ATM pull v1 ATM pull v2 ATM pull v3 Bank deposit v1 Bank deposit v2 Bank deposit v3 Request for statement v1 Request for statement v2 Request for statement v3
- 25. Undoing changes Original event State is OK Problem event(s) Events are added that put the system in an error state Changes New events are impacted Re-add original state Read the original state Reapply changes
- 26. Patterns of use
- 27. Projections
- 29. Map/Transformation Event v1 Event v2 Event v3 Event v4 Request event Rehydration transform Business logic transform Response event
- 30. Dispatcher
- 31. Throttle
- 32. State machine
- 33. Workflow
- 34. Composability
- 35. Retry pipeline Event Send fail event Original event Transform Increment try count on original event State machine If try count > max tries => dump If try count < max tries => send Dispatcher Send message Log message send attempt
- 36. Load balancer State machine Use information from the events to decide where they go Mark events accordingly Dispatcher Send events based on markings made by state machine
- 38. Interacting with outside systems Time bound values (i.e. money) Systems where reads have side effects (i.e. credit scores) Log everything that you send
- 41. Apache Kafka
- 43. Reading/watching list Article about event sourcing with some code samples https://martinfowler.com/eaaDev/EventSourcing.h tml Presentation about event sourcing https://www.youtube.com/watch?v=JHGkaShoyNs Presentation about source control as data https://www.youtube.com/watch?v=KaLROwp-VDY
- 44. Call to action