Why build resource-hungry software: principles of the reconciliation loop (hello, K8s!)
summary
Explains why building resource-hungry (“greedy”) software can be beneficial and describes the reconciliation loop pattern used by systems like Kubernetes. Covers problems of managing large distributed systems with partially uncontrollable subsystems and shows how continuous reconciliation (observe–compare–act cycles) helps detect and correct drift between desired and actual state.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 3:35Control vs Data Plane
- 8:57Error handling
- 12:51Bidirectional comms
- 19:11Ideal architecture
- 22:43Reconciliation loop
- 29:47Pros and cons
- 34:21Summary
- 35:45Q&A
related talks
35mBeyond Reconcile Loop - Patterns for Kubernetes Operator Development
36mUnderstanding Kubernetes Through Real-World Phenomena and Analogies
31mIt's all about reconciliation - anatomy of a kubernetes controller
26mThe Control Loop As An Application Development Framework
01h 27mTutorial: How To Write a Reconciler Using K8s Controller-Runtime!
39mBuilding Distributed Systems with Kubernetes
41mA Defence of Technical Excellence
01h 02mTools for simplified large scale systems