State of Affairs or Affairs of State
summary
Cloud services move from stateless round‑trips to keeping entity state in memory for performance and cost savings; Orleans implements virtual actor (grain) objects with stable identities, single‑threaded activations, load‑on‑demand and passivation to limit DB trips, and lets grains own write‑through or batch persistence. Three failure patterns are compared: immediate failure return, persistent queues, and durable workflows. Temporal is workflow‑first code‑first: it persists workflow state, schedules retries, and uses workers to run activities, outsourcing orchestration for durable retries and long‑running state.
AI-written summary · about
chapters (18)
- 0:00Intro
- 0:24Stateless vs Stateful
- 1:12Whiteboard
- 2:50Stateless Architecture
- 9:34Release
- 13:50Grains
- 16:18Handling failures
- 16:55Request response
- 19:00Tradeoffs
- 19:20Persistent Queue
- 21:17Pros and Cons
- 21:29Workflow Approach
- 23:16Benefits of Workflow Approach
- 24:56What is Temporal
- 27:22Temporal Workflow
- 30:26Conclusion
- 31:11Summary
- 32:11Questions
related talks
01h 04mOrleans Architecture Patterns
28mStateful Applications in the Age of Durable Execution
01h 01mHow to build real-world applications with Orleans
31mBuilding Asynchronous SOA for Modern Apps
42mHow to build distributed systems in JavaScript
26mMicroservices Madness: Application And Infrastructure Automation Using Temporal
36mLost In Transaction
46mLost in transaction? Strategies to deal with (in-)consistency in distributed systems