Keynote: Introducing Stateful Functions 2.0: Stream Processing meets Serverless Applications
summary
Stateful Functions 2.0 separates compute from state so Flink keeps the state and messaging while user logic runs as remote functions (Kubernetes deployments or FaaS like AWS Lambda), preserving exactly‑once semantics and consistent messaging. Demo shows a Python ML model scaled independently on k8s with zero‑downtime rollouts and instant fixes. Release also adds container memory management and RocksDB integration plus experimental underlined checkpoints to reduce checkpoint latency, and promotes a language‑agnostic SDK model.
AI-written summary · about
chapters (15)
- 0:00Introduction and welcome
- 1:08Apache Flink community updates
- 2:55Streaming SQL and ecosystem
- 4:42SQL integration demo
- 7:38Python support in Flink
- 9:05Memory management improvements
- 12:15Underlined checkpoints explained
- 15:47Stateful Functions overview
- 20:23Goals for Stateful Functions 2.0
- 22:56Streaming vs database architecture
- 30:29Stateful Functions 2.0 architecture
- 32:38Serverless applications with stateful functions
- 35:10Machine learning demo
- 39:53Operational benefits demo
- 40:14Contributing and closing
related talks
34mHands-on Stateful Serverless Applications with Kubernetes and Stateful Functions
27mStateful Functions 3.0: Towards a Cross-Platform Stateful Serverless Stack
35mHands-On Stateful Serverless Applications with K8s and Stateful Functions
38mDeploying Stateful FaaS on Streaming Dataflows
25mCloudstate—Towards Stateful Serverless
13mKeynote: What Next? Expanding Knative and Serverless
35mServerless is Interesting But FaaS is Not Enough
40mThe State of Serverless Computing