Drinking a River of IoT Data with Akka.NET

on YouTube ↗52m

" summary

Shows an Akka.NET-based IoT backend that ingests meter readings, normalizes timestamps, interpolates gaps and computes consumption; uses Akka.Remote for proxies to route messages and Akka.Persistence (event sourcing + snapshots) for a single hot actor per device with cold offload to DB. Claims up to 50 million messages/sec but reports 2 million messages/sec on an aging laptop. Describes a gap-fill using the previous week’s trend, supervision/restart strategies, and a working cluster workflow for redeploy and recovery.

AI-written summary · about

" chapters (14)

1. 0:00Intro
2. 0:23Agenda
3. 0:55A bit of history
4. 8:00Introduction to Akka.NET
5. 19:19The problem domain
6. 22:39How Akka.NET fits in
7. 26:20Implementation details
8. 27:00Normalizing measurements
9. 33:47Getting messages to the ActorSystem
10. 39:33Persisting data
11. 47:00Restart behavior
12. 49:23Beyond this talk
13. 50:52Conclusion
14. 51:17Outro

" related talks

• 59mDrinking a river of IoT data with Akka.NETHannes Lowette · Build Stuff 2019
• 48mProcess IoT information using Akka.NETAarjan Meirink · Build Stuff 2017
• 50mActors can rule your DDD worldHannes Lowette · KanDDDinsky 2022
• 01h 14mAkka.NET: The Future of Distributed Programming in .NETAaron Stannard · NDC Sydney 2016
• 39mActors can rule your DDD worldHannes Lowette · Craft Conference 2022
• 31mFun with SensorsYogi Rampuria · Distributed SQL Summit Asia 2022
• 20mBuilding realtime efficient queries in an IOT platformDinesh Narayanan · Data By the Bay 2016
• 55mCode it like C#, run it like ErlangPhilip Laureano · NDC Sydney 2016