[ https://issues.apache.org/jira/browse/IGNITE-10876?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16757382#comment-16757382
]
Sergey Kosarev commented on IGNITE-10876:
-----------------------------------------
Overall changes look good to me.
I'd suggest to create shortcut doInParallel method to decrease copy-pastes of parllelism and
systemExecutorservice, but it's optional.
> "Affinity changes (coordinator) applied" can be executed in parallel
> --------------------------------------------------------------------
>
> Key: IGNITE-10876
> URL: https://issues.apache.org/jira/browse/IGNITE-10876
> Project: Ignite
> Issue Type: Improvement
> Reporter: Pavel Voronkin
> Assignee: Pavel Voronkin
> Priority: Major
> Time Spent: 0.5h
> Remaining Estimate: 0h
>
> There is for loop over all cache groups which execution N*P operations in exchange worker
where N is number of cache groups, P is number of partitions.
> We spend 80% of time in a loop
> for (CacheGroupContext grp : cctx.cache().cacheGroups()){
> GridDhtPartitionTopology top = grp != null ? grp.topology() : cctx.exchange().clientTopology(grp.groupId(),
events().discoveryCache());
> top.beforeExchange(this, true, true);
> }
> I believe we can execute it in parallel
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
|