[
https://issues.apache.org/jira/browse/IGNITE-12719?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sunny Chan updated IGNITE-12719:
--------------------------------
Environment: (was: We are running Ignite cluster on bare metal on a
relatively high core count machine (4x10 cores 20 threads), and looking some of
the thread pool initialization code:
{{(IgnitionEx.java)}}
{{sysExecSvc = *new* IgniteThreadPoolExecutor(}}{{
"sys", }}{{cfg.getIgniteInstanceName(),}}{{
cfg.getSystemThreadPoolSize(),}}{{
cfg.getSystemThreadPoolSize(),}}{{
*_DFLT_THREAD_KEEP_ALIVE_TIME_*, }}{{*new* LinkedBlockingQueue<Runnable>(),}}{{
GridIoPolicy.*_SYSTEM_POOL_*);}}
Notice that the core thread pool size is equals to the max thread pool
settings, which is by default same as the number of CPU cores. And in our
cases, we won’t be reusing any threads until we have enough request coming in
to fill 80 threads. Also, we might want to tune the thread keep alive time to
improve thread reuse.
We would like to propose to change ignite so that users can configure the core
thread pool size in these Ignite thread pools.)
> Allow users to configuring Ignite Thread Pool's Core thread count/max thread
> count/etc
> --------------------------------------------------------------------------------------
>
> Key: IGNITE-12719
> URL: https://issues.apache.org/jira/browse/IGNITE-12719
> Project: Ignite
> Issue Type: Bug
> Reporter: Sunny Chan
> Priority: Major
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
