Fog Computing and IoT Solution Architecture
summary
Fog computing claim: run cloud-like microservices at the edge using Eclipse io fog with a distributed data bus that spools to disk and an in-memory layer achieving about 15 microseconds latency at a million msgs/sec on decent hardware. Demos use Python TensorFlow image recognition to shrink a 1 MB photo to ~200 bytes, a public pipe that opens a secure URL into a room, and Bluetooth Low Energy adapters on Raspberry Pi zero (≈6 microservices) up to Pi3 (≈12) and Digi gateways (≈24); heart‑rate → Epic/Hadoop integration proves it works.
AI-written summary · about
chapters (12)
- 0:00Start
- 4:11Machine-to-Machine (M2M)
- 5:27M2M meets Internet
- 17:57The new landscape
- 18:30What's being done right now
- 20:29What is fog computing?
- 22:52Fog is VERY different from cloud
- 32:13Business reasons for fog computing
- 41:00loT solution architecture concerns
- 42:32Some Eclipse ioFog hardware choices
- 46:07Edge analytics with Google TensorFlow
- 50:20Patient heart rate with HIPAA compliance
related talks
53mFull-Scale Edge Computing with Eclipse ioFog: A Case Study
28mDistributed IoT Applications with
39mExploring the Trend Toward the Edge
20mFrom Cloud to Fog Computing and IoT
41mCloud Computing: The Edge talks - long road into the Cloud
53mFog Computing : Software on the Edge
29mLife on the Edge
01h 02mIoT on the Edge