FaaS is Not Only the Serverless: Stream Processing with Serverless
summary
Claims a serverless model that runs container per function for long-lived WebRTC streams and spawns containers per session so each stream gets isolated, state-free functions. Built on Kubernetes with gRPC + WebRTC and GStreamer, it demonstrates voice sent from a browser, cascaded speech-to-text and translation using Google Speech API, and a live demo. It creates over 100 pods in minutes, manages lifecycle via a custom controller, propagates trace span context into containers, and publishes code on GitHub.
AI-written summary · about
chapters (7)
- 0:00Start
- 3:22Voice from customers
- 4:06Nend cloud computing power and
- 5:39Our Descision
- 25:23Cloudevents
- 26:09Telemetry
- 27:10Recap
related talks
34mFaaS and Furious - 0 to Serverless in 60 Seconds, Anywhere
23mOpenFaaS: Serverless platform with no strings attached?
38mServerless with OpenFaaS - Live Example
33mFaaS 101
51mFaaS or not to FaaS? Visible and invisible benefits of the Serverless paradigma
19mDiscovering serverless with OpenFaaS
30mServerless State of the Union
19mpgFaaS: Geo-Processing made simple with Serverless technology
