[
https://issues.apache.org/jira/browse/IGNITE-4106?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15629128#comment-15629128
]
Andrew Mashenkov commented on IGNITE-4106:
------------------------------------------
Approach to split sql query by sending Map queries messages with subset of
primary partitions is seems to be more simple.
Although it mechanics similar to multi-node, I would expect equals trend of
throughput as for threads count as nodes count.
I've no success to find a reason yet, it seems there are contentions somewhere
and I don't have enough expertise yet to find it alone.
> SQL: parallelize sql queries over cache local partitions
> --------------------------------------------------------
>
> Key: IGNITE-4106
> URL: https://issues.apache.org/jira/browse/IGNITE-4106
> Project: Ignite
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 1.6, 1.7
> Reporter: Andrew Mashenkov
> Assignee: Andrew Mashenkov
> Labels: performance
>
> If we run SQL query on cache partitioned over several cluster nodes, it will
> be split into several queries running in parallel. But really we will have
> one thread per query on each node.
> So, for now, to improve SQL query performance we need to run more Ignite
> instances or split caches manually.
> It seems to be better to split local SQL queries over cache partitions, so we
> would be able to parallelize SQL query on every single node and utilize CPU
> more efficiently.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
