Introducing Flink on Mesos
summary
Flink gains native Mesos deployment in the Flink 1.2 line with a working demo that launches a Mesos session producing a jobmanager and taskmanager pair and successfully runs a WordCount job. Resource packing uses the Netflix FENsO optimizer to solve the knapsack placement problem across fragmented offers. A long‑running dispatcher for multi‑cluster management exposes a REST gateway so CLI can spawn persistent clusters; Docker and local disk reservation are planned, and dynamic within‑job scaling is in progress.
AI-written summary · about
chapters (10)
- 0:00Introduction to Flink on Mesos
- 3:56Why run Flink on Mesos
- 8:51First Flink job on Mesos demo
- 10:16How Mesos integration works
- 15:40Optimizing task placement
- 22:56Role of the Flink dispatcher
- 29:43Elastic Flink clusters on demand
- 32:52Future of Flink on Mesos
- 37:09Comparison with Kubernetes
- 40:47Q&A and closing
related talks
43mApache Flink Meets Apache Mesos and DC/OS
30mDemystifying Deployments: Applications or Clusters, Active and Reactive Scaling - What is it about?
46mFlink SQL in 2020: Time to show off!
40mFrom concept to production; the fastest way to develop & deploy Apache Flink jobs
42mApache Flink deep-dive #bbuzz
22mDo Flink on Web with FLOW
36mApache Flink in the Cloud-native Era
42mManaging Apache Flink on Kubernetes - FlinkK8sOperator
