[
https://issues.apache.org/jira/browse/KYLIN-4985?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17370399#comment-17370399
]
tianhui commented on KYLIN-4985:
--------------------------------
Hi [~yaho], I'm sure *kylin.cube.cubeplanner.expansion-threshold* can control
redundancy, but is it better to just delete unhitted cuboids in variable space?
Why Kylin should prefer to build parent cuboid rather than the hitted cuboid
itself?
As for weighting change, It's just a hueristic change to punish the cuboid that
hitted less. Because it looks like the *hitProbability* in CuboidBenefitModel
has never been used, I think it's better to consider hit probablility in
cuboids recommand.
> optimize kylin planner by delete unnecessary cuboids
> ----------------------------------------------------
>
> Key: KYLIN-4985
> URL: https://issues.apache.org/jira/browse/KYLIN-4985
> Project: Kylin
> Issue Type: New Feature
> Reporter: tianhui
> Priority: Major
>
> When I use Kylin Planner, I can get many cuboids in recommand result that
> never hitted by my history queries. I think it maybe unnecessary, so I delete
> the unhitted cuboids.
> In addition, I change row count by weighting of 1/sqrt(hit probability)
> before execute plan algorithm.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
