Introduction to Event Sourcing
Download as PPTX, PDF0 likes1,154 views
The document outlines a presentation by Jeffrey T. Fritz on event sourcing, an architectural pattern for managing entity state through the storage of immutable events rather than traditional methods. It discusses the benefits and challenges of implementing event sourcing, including performance metrics and various data access technologies. The presentation also highlights tools for event storage, reporting requirements, and testing methodologies related to this architectural approach.
Introduction to Event Sourcing
- 1. Jeffrey T. Fritz @csharpfritz http://csharpfritz.com
- 2. Jeffrey T. Fritz ASP.Net Developer Evangelist for Telerik AJAX controls Microsoft MVP, ASPInsider Tech Ed 2012 Speaker Idol Finalist, 2013 Runner-up 15 years of web development experience CodeProject.tv + Pluralsight author (coming soon!) May 2013 MSDN Magazine: One ASP.Net DotNetRocks #855 Yet Another Podcast #98, #106
- 3. Level setting: n-tier architecture Event Sourcing – definition, benefits, challenges How do I get started? How do I audit? Rebuilding and Reporting Samples
- 4. Presentation Layer Business Logic Layer Data Access Layer Database
- 5. Event Sourcing: An architectural pattern which warrants that your entities do not track their internal state by means of direct serialization or O/R mapping, but by means of reading and committing events to an event store Event: an immutable object that describes a change that has occurred to an entity.
- 6. Entities are not stored as a given state in time Changes to the entities are persisted Stored in the order changes were applied to the object No concern for ‘prior state’ of the entity
- 7. Image courtesy of Greg Young
- 11. Simplified data access ◦ Append-only data store ◦ Easy to partition ◦ Easy to replicate Event structure allows for greater use of ‘ubiquitous language’ Reporting requirements are required to be stored separately
- 12. Performance ◦ Documented performance on an i7 laptop with SSD storage: 15,000 writes / second 50,000 reads / second ◦ Aggressively cache query results Data never changes
- 13. Testing ◦ Given (Collection of Events) ◦ When (Action) ◦ Then (Events expected) No duplicate storage of data ◦ Current State ◦ Audit log
- 14. Tivo for your entities ◦ Trivial to ‘replay’ the events to a given point in time Offline processing Conflict Management
- 15. Not storing data in relational structure ◦ Can use RDBMS as a store Reporting is NOT immediately available Difficult to migrate a legacy app to Harder to sell to an organization
- 16. ObjectID Change Date Change User Field Old Value New Value 1 9/20/12 Jeff Attend 0 1 ObjectID Version DateStamp Event 1 5 9/20/12 RsvpChanged { Name: Jeff, Attending: true }
- 17. Jonathan Oliver EventStore ◦ .Net based tool ◦ Supports storage in: MySQL, MSSql, SQLite, Access, Sql Azure, Amazon RDS, Oracle, Sybase, Postgres, DB2, RavenDb, MongoDb, CouchDb, and LOTS more ◦ http://github.com/joliver/EventStore Greg Young Event Store ◦ JSON based event store ◦ Proprietary storage engine ◦ RESTful access ◦ http://geteventstore.com
- 18. Demo: Jonathan Oliver Event Store Get and Save
- 19. AuditLog is just a dump of events Call ToString on each event Demo – NerdDinner AuditLog
- 20. Data can be “re-projected” into any format New data can be discovered by the projection that wouldn’t normally be visible in 3rd normal form Alternate storage of data in format for the report desired
- 21. Offline re-build of reporting data NerdDinner.Rebuild
- 22. Use AAA syntax ◦ Arrange (collection of events) ◦ Act (action) ◦ Assert (verify output of events)
- 23. http://bit.ly/FritzES -- Event Sourcing Links http://csharpfritz.com @csharpfritz on Twitter Thanks for attending!