Building Event Streaming Applications with Serverless Kafka
summary
A Pac‑Man game is deployed end‑to‑end to demonstrate event streaming: a Terraform deployment takes about four to five minutes to create an S3‑hosted frontend, AWS API Gateway + Java Lambda that produces events into Confluent Cloud Kafka, ksqlDB (running on ECS/Fargate) performs aggregations into a scoreboard topic, and a Go consumer subscribes to show a live table. Demo works live; main takeaway is that serverless APIs are easy but statefulness (materializing aggregations) forces architectural tradeoffs.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 1:57Objectives and setup
- 4:20Architecture and deployment
- 8:06Configuration and plan
- 10:43Component walkthrough
- 15:25Interactive demo
- 17:52KSQLDB and data processing
- 24:07Serverless challenges
- 25:58Q&A and conclusion
related talks
12mServerless Stream Processing with Confluent Cloud, Kafka and AWS Lambda
36mLeveraging Serverless architecture with Confluent Kafka
32mBuilding and scaling a log analytics platform - a serverless approach
28mThe Future of Streaming: Global Apps, Event Stores and Serverless
47mServerless Kafka Patterns.
38mKafka on the Fly: A Serverless Approach to Data Streaming
35mStreaming and Real-Time Analytics
29mHow to build a social network on serverless