[jira] [Resolved] (KUDU-3407) MM: Give a chance to do other OP while server is under memory pressure

["Alexey Serbin (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/KUDU-3407?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Alexey Serbin resolved KUDU-3407.
---------------------------------
    Fix Version/s: 1.18.0
       Resolution: Fixed

This is addressed with the changelist described in the [previous comment|https://issues.apache.org/jira/browse/KUDU-3407?focusedCommentId=17788142&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-17788142].  To use the new functionality, set {{\-\-run_non_memory_ops_prob}} flag to a value greater than 0 and less than 1.

It's a separate question whether the behavior of the maintenance manager should be changed by setting the new flag to a non-zero value by default.  More comprehensive analysis and testing is required to find a good default setting for the flag -- feel free to open a new JIRA item to address that.

> MM: Give a chance to do other OP while server is under memory pressure
> ----------------------------------------------------------------------
>
>                 Key: KUDU-3407
>                 URL: https://issues.apache.org/jira/browse/KUDU-3407
>             Project: Kudu
>          Issue Type: Improvement
>          Components: compaction
>    Affects Versions: 1.14.0
>            Reporter: Song Jiacheng
>            Priority: Major
>             Fix For: 1.18.0
>
>
> For now, if the server is under pressure(60% memory as default), MaintenanceManager always find a memory-flush operation to run. So if all the tservers are under pressure for a long time, there will be many pending operations, which will lead to many problem, like redos not compact, undo not delete, etc..
> It might be better if we add a parameter to give a chance to do other ops, the formula is like this:
> Do other works? = (1 - (memory_now - pressure_threshold) / (soft_limit - pressure_threshold)) * kChance
> kChange should be configurable.
> This will give a probability to run other maintenance operations.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)