The Future of Streaming: Global Apps, Event Stores and Serverless
summary
Argues that Apache Kafka powers global streaming with clusters of hundreds (some installations pushing thousands) and topics holding hundreds of terabytes, enabling event storage that can rewind and replay histories for reprocessing. Gives concrete systems: a Formula One truck running a small Kafka cluster ingesting ~400 sensors per car, and the New York Times using a single “monologue” topic (they cite 1851 items) as a canonical stream. Contrasts stateful stream processors (transactions, materialized tables) with short‑lived serverless functions that scale to thousands.
AI-written summary · about
chapters (7)
- 0:00Introduction to streaming
- 3:38Event-driven architectures
- 9:57Event sourcing explained
- 13:51Formula One streaming case
- 17:05Serverless and streaming
- 23:49Stream processing vs FAAS
- 26:36Summary of streaming concepts
related talks
28mBuilding Event Streaming Applications with Serverless Kafka
38mKafka on the Fly: A Serverless Approach to Data Streaming
12mServerless Stream Processing with Confluent Cloud, Kafka and AWS Lambda
56mScaling Telemetry Systems with Streaming & Serverless
20mUnleashing the Power of Serverless Architecture
24mShaping Tomorrow's Technology: Navigating Cloud-Native, Serverless, and Polyglot Programming
36mLeveraging Serverless architecture with Confluent Kafka
47mServerless Kafka Patterns.