Lets Deeploy Flink - Uncovering Hidden Depths of Yarn, Docker, Kubernetes & Beyond
summary
Compares concrete Flink deployment choices—Standalone, Docker, Kubernetes, YARN—with how each manages artifacts, HA, scaling and savepoints and when to use session vs application modes explained. Lists Kubernetes options—standalone YAML, native CLI, and operator—and claims 90–95% Kubernetes usage among customers, showing operator benefits like declarative CRs, CI/CD fit and automated recovery. Demonstrates artifact strategies (init container, image, S3), resource tradeoffs and that the operator automates savepoint and lifecycle so upgrades and autoscaling work without manual steps.
AI-written summary · about
chapters (13)
- 0:00Apache Flink introduction
- 1:52Flink app types and APIs
- 5:25Session vs application modes
- 9:36Managing state and artifacts
- 12:04Mini and local clusters
- 13:50Local cluster with YARN
- 18:02Flink on Hadoop YARN
- 21:32Docker deployments
- 23:41Kubernetes standalone
- 29:00Native Kubernetes Flink
- 32:40Kubernetes operator
- 39:06Deployment options comparison
- 41:33Flink deployment best practices
related talks
26mKubernetes & real-time analytics: How to connect these worlds with Apache Flink
22mBuilding a Kubernetes Operator for Apache Flink in Java
33mFlink in Containerland
15mFlink's New Dockerfile: One File to Rule Them All
37mSpark and Flink Running Scalable in Kubernetes #bbuzz
36mRunning Flink in production: The good, the bad and the in-between
30mCloudflow: Spark, Flink, and Akka Working together on Kubernetes
58mBuilding Self Service Platforms for Apache Flink