[
https://issues.apache.org/jira/browse/HBASE-17788?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15940783#comment-15940783
]
stack commented on HBASE-17788:
-------------------------------
Fantastic [~gubjanos] Thanks for the help.
Make yourself an issue and link it here. I added you as a contributor so you
should be able to hack around in JIRA.
bq. Just to clarify, is this about the WorkerThreads in ProcedureExecutor.java?
Because the construction there is somewhat similar to a ThreadPool Executor.
That is my understanding. They run the procedures. What you thinking? Replace
w/ ThreadPool Executor? That sounds good. There is means of monitoring for
stuck threads/procedures. We need that. Reporting on what is stuck could be
better; IIRC, it just says 'stuck' w/o saying which procedure is stopped.
Thanks again.
> Procedure V2 performance improvements
> -------------------------------------
>
> Key: HBASE-17788
> URL: https://issues.apache.org/jira/browse/HBASE-17788
> Project: HBase
> Issue Type: Sub-task
> Components: Performance, proc-v2
> Reporter: stack
> Fix For: 2.0.0
>
>
> This is a list of items to work on to improve framework perf taken from
> https://docs.google.com/document/d/1kEWzyA0iCyRjdogjju9JNMDT9ODHaKcyoAJTMXkDpxc/edit#
> Make sub-issues to work on each in the list below:
> * Replace fixed Executor Threads with dynamic thread pool (Easy)
> * Connect the active thread count with the ProcStore slots (Medium)
> * Parallelize the load (Complex)
> * Allow Procedures to start early (Complex)
> https://github.com/apache/hbase/blob/master/hbase-procedure/src/main/java/org/apache/hadoop/hbase/procedure2/store/wal/ProcedureWALFormatReader.java#L155
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
