[FLINK-3322] MemoryManager creates too much GC pressure with iterative jobs

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Abandoned
• Affects Version/s: 1.0.0
• Component/s: Runtime / Task
• Labels: pull-request-available
• Assignee: ramkrishna.s.vasudevan
• Reporter: Gábor Gévay
• Created: 02/Feb/16 15:26
• Updated: 14/Oct/19 11:33
• Resolved: 14/Oct/19 11:25

Description

When taskmanager.memory.preallocate is false (the default), released memory segments are not added to a pool, but the GC is expected to take care of them. This puts too much pressure on the GC with iterative jobs, where the operators reallocate all memory at every superstep.

See the following discussion on the mailing list:
http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/Memory-manager-behavior-in-iterative-jobs-tt10066.html

Reproducing the issue:
https://github.com/ggevay/flink/tree/MemoryManager-crazy-gc
The class to start is malom.Solver. If you increase the memory given to the JVM from 1 to 50 GB, performance gradually degrades by more than 10 times. (It will generate some lookuptables to /tmp on first run for a few minutes.) (I think the slowdown might also depend somewhat on taskmanager.memory.fraction, because more unused non-managed memory results in rarer GCs.)

Issue Links

• links to: GitHub Pull Request #1769
• links to: GitHub Pull Request #2495
• links to: GitHub Pull Request #2495
• links to: GitHub Pull Request #2510
• links to: GitHub Pull Request #2510