Why should you care about edge computing?
summary
Explains when moving compute, storage and decision logic onto devices pays off: privacy-preserving on-device facial models for doorbells and hospitals, a CNC site cutting failure detection from 2 s to 100 ms, and a bespoke rig ingesting ~3 GB/s (≈300 TB/day) with GPUs for FFTs. Shows concrete tooling: Azure IoT Edge and containers of Cognitive Services, AWS Greengrass, Snowball/Snowmobile for remote ingestion, examples of missing OTA updates across 60,000 devices, and realities: bandwidth costs, security risks (botnet fridges) and maintenance trade-offs.
AI-written summary · about
chapters (34)
- 0:00Intro
- 0:52Expectations
- 1:54Gartner Hype Cycle
- 5:15Internet of Things
- 5:57Ring Security Camera
- 6:56Data Centers
- 7:36Smart Devices
- 8:34Edge Devices
- 10:52Privacy
- 11:58Amazon patent
- 15:37Image recognition
- 19:06Healthcare
- 20:21Latency
- 22:50Voice Recognition
- 23:34Resistance to Failure
- 25:41Cloud is gone
- 26:12Size
- 26:39Data analysis
- 29:22Bandwidth
- 30:21Connected devices
- 32:05Security
- 35:05Your Sphere
- 36:18Azure IoT Hub
- 36:53AWS Free RTOS
- 37:49Physical Security
- 39:08Edge Computing
- 40:33Cognitive Services
- 43:56IOT Edge
- 47:27AWS Greengrass
- 49:02Snowball
- 50:55Google Tensorflow
- 52:42Azure Kinect Camera
- 54:37Edge Computing Devices
- 55:55Conclusion
related talks
49mEnable IoT with Edge Computing and Machine Learning
20mHow ML is Powering the Future of IoT
31mBuild Your Intelligent Edge
29mEdge Intelligence Revolution
13mHow Edge AI Will Save The World
21mA New Architecture for the Intelligent Edge
22mEdge AI Use Cases in IoT — Unlocking Intelligence at the Network Edge
01mWhy edge AI is becoming the default for devices