The Brave New World of serverless computing
summary
Argues serverless excels for one-off jobs, webhooks and glue logic tying SaaS together, not for heavy ML training or custom hardware; shows a Node.Js demo using Pusher Channels plus Google Cloud Speech-to-Text and static hosting (Netlify/now) that streams captions and scales to thousands with zero ops. Notes tradeoffs: per-invocation cost is higher than EC2 but operational savings speed shipping, and tooling like Auth0, Glitch, GitHub Actions and hosted Kafka make composition trivial.
AI-written summary · about
chapters (17)
- 0:00Intro
- 0:54History of shipping
- 4:59Service everything
- 6:36Functions
- 9:27Limitations
- 11:43Glitch
- 15:17SAS
- 15:50Oddzero
- 16:47Kafka
- 18:00Nataljaphi
- 18:54Sight
- 21:00Serverless is inevitable
- 21:55Summary
- 22:35Project
- 23:29Streaming
- 25:58Is it serverless
- 26:57Whats next
related talks
58mServerless - the brief past, the bewildering present, and the beautiful (?) future
35mServerless Experiences -- On Your Own Servers?
42mNo Server Is Better Than No Server: Building Blocks for Serverless Architectures
28mServerless - the way microservices should be
01h 00mServerless - reality or BS - notes from the trenches
29mGoing Serverless
32mEmbracing Serverless Ops
28mFun With Serverless Javascript