Implementing an Event Sourcing strategy on Azure
summary
Concrete recipe: producers push JSON events into Event Hubs (Kafka endpoint), an Azure Function reads the stream and inserts documents into Cosmos DB as an append‑only event store, then another Function consumes the Cosmos DB change feed to build cheap point‑read projections (warehouse/day and customer views). Demo sends 1,000 orders × ~5 events each, everything lands in Cosmos and projections update; functions and Event Hubs are serverless so latency is milliseconds and cost is driven by RU consumption, with a free Cosmos tier available for small tests.
AI-written summary · about
chapters (23)
- 0:00Start
- 1:50Introduction
- 2:02What Is Event Sourcing
- 2:25Create Events
- 3:51Event Sourcing
- 4:03Why Should We Use Event Sourcing
- 6:02The Event Sourcing Pattern
- 8:05Event Sourcing on Azure
- 8:28Event Sourcing and Azure
- 10:08Event Hubs
- 11:18Azure Functions
- 12:11Event Sourcing Pattern on Azure
- 17:48Cosmos Db
- 21:40Event Sourcing Pattern
- 26:02Azure Functions with the Trigger
- 28:43Demo
- 34:04The Custom Logic for Creation and Update of Document
- 38:18Key Takeaways
- 39:50Benefits
- 40:41The Ecosystem
- 46:56Cost Calculator
- 54:13When Would You Use the Cosmos Db
- 55:29Close Session
related talks
42mImplementing an Event Sourcing Strategy on Azure
46mPractical CQRS and Event Sourcing on Azure
52mEvent Sourcing – the what, why & how
57mEvent driven architectures on Azure
01h 01mMessage Brokering and Event Driven Architecture on Azure
45mCommon quandaries implementing your first Event Driven Architecture
52mEvent Sourcing for the Cloud Developer
26mImplementing Event-Driven Microservices Architecture (...)