Meshing Robots and Humans with Distributed Apps
summary
Factories need an eventually consistent decentralized log that lets sensors and tablets be always available for write (local buffering during Wi‑Fi gaps) and later sync; uses actor-style processing, Lamport clocks or timestamps to get causal ordering, snapshots for fast replay, and optional consensus for strict constraints. Example builds: a temperature stream that waits five samples then emits an alarm, and a decider emits sent event to record SMS delivery — duplicates occur if two redundant deciders act before seeing each other’s events.
AI-written summary · about
chapters (9)
- 0:00Factory automation software
- 3:06How factories operate
- 9:40Computation and communication
- 17:49Factory event log requirements
- 26:45Event ordering and consistency
- 34:15Transforming event streams
- 41:10Decisions in distributed systems
- 47:13Distributed databases and serializability
- 58:13Conclusion and future directions
related talks
26mLocal-first
01h 03mOffline first computing in an online world
57mFrom Shop Floor Control to Physical AI: Bridging Industrial Connectivity and Intelligent Automation
43mAn Approach For The Internet Of Things
38mEventual Consistency: A Question of Time and Expectations
17mEnsuring Reliable Distributed Transactions in Logistics
32mEvent-Driven Infrastructure: Autonomous Operation and Unreliable Networks
39mTrust Deterministic Execution to Scale & Simplify Your Systems