[GitHub] [incubator-doris] chenhao7253886 opened a new issue #1053: New query scheduler

[GitBox <gi...@apache.org>]

chenhao7253886 opened a new issue #1053: New query scheduler
URL: https://github.com/apache/incubator-doris/issues/1053
 
 
   Now when query plan fragments are distributed to BE, the load of BE is not considered. occasional big queries maybe cause some BE overload, because we only randomly choose BE, so some overload BE maybe selected and the load of BE is more heavily, as last it cause all system's query to hang for dozens of minutes or even a few hours. so i want to solve this with new scheduler and recover query running at minutes level. 
   I expect to add some simple strategies to new query scheduler, when distributing plan fragment the BE which overload for several minutes can be excluded.  the following system metrics can be collected for this:
   1. the ioutil In the last few minutes
   2. the mem unused In the last few minutes
   and the system metrics only work in a few minutes.
    I will analyze some user's cases and conform this can work.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@doris.apache.org
For additional commands, e-mail: dev-help@doris.apache.org