[FLINK-12122] Spread out tasks evenly across all available registered TaskManagers

• Type: Sub-task
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.6.4, 1.7.2, 1.8.0
• Fix Version/s: 1.9.2, 1.10.0
• Component/s: Runtime / Coordination
• Labels: pull-request-available
• Release Note: In order to use a scheduling strategy which is more similar to the pre FLIP-6 behaviour where Flink tries to spread out the workload across all available `TaskExecutors`, one needs to set `cluster.evenly-spread-out-slots: true` in the `flink-conf.yaml`
• Assignee: Till Rohrmann
• Reporter: Till Rohrmann
• Created: 08/Apr/19 09:39
• Updated: 31/Jul/20 10:14
• Resolved: 07/Nov/19 22:40

Description

With Flip-6, we changed the default behaviour how slots are assigned to TaskManagers. Instead of evenly spreading it out over all registered TaskManagers, we randomly pick slots from TaskManagers with a tendency to first fill up a TM before using another one. This is a regression wrt the pre Flip-6 code.

I suggest to change the behaviour so that we try to evenly distribute slots across all available TaskManagers by considering how many of their slots are already allocated.

Issue Links

• causes: FLINK-15013 Flink (on YARN) sometimes needs too many slots
• is duplicated by: FLINK-11148 Rescaling operator regression in Flink 1.7
• links to: GitHub Pull Request #9928
• links to: GitHub Pull Request #9929 backport to 1.9