You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Zhu Zhu (Jira)" <ji...@apache.org> on 2019/11/27 12:32:00 UTC

[jira] [Created] (FLINK-14974) Managed memory fractions should be rounded properly to not summed up to be more than 1.0

Zhu Zhu created FLINK-14974:
-------------------------------

             Summary: Managed memory fractions should be rounded properly to not summed up to be more than 1.0
                 Key: FLINK-14974
                 URL: https://issues.apache.org/jira/browse/FLINK-14974
             Project: Flink
          Issue Type: Sub-task
          Components: Runtime / Coordination
    Affects Versions: 1.10.0
            Reporter: Zhu Zhu
             Fix For: 1.10.0


Managed memory fractions should be rounded to floor at a certain precision when divided by the number of operators, otherwise the fractions can be summed up to be more than 1.0 due to the double precision issue, and the last operator may fail to allocate managed memory it is supposed to be able to acquire.

To achieve that, I think we should change {{StreamingJobGraphGenerator#setManagedMemoryFractionForOperator}} to use BigDecimal#divide(otherValue, scale, RoundMode.ROUND_DOWN) to calculate the fractions. In this way, the sum of the fractions will not exceed 1.0.
The scale can be a bit larger (maybe 16) so that we only lose little managed memory.

cc [~chesnay]



--
This message was sent by Atlassian Jira
(v8.3.4#803005)