[jira] [Commented] (CALCITE-3657) EnumerableHashJoin should not use NLogN for costing

["Ruben Q L (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CALCITE-3657?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17006806#comment-17006806 ] 

Ruben Q L commented on CALCITE-3657:
------------------------------------

Since this ticket is modifying the computeSelfCost of EnumerableHashJoin, I think we could also introduce another factor to make the cost computation more accurate: the (hash)join condition.
With the implementation of CALCITE-2973, now HashJoin supports all types of join condition, so we can have three different scenarios:
a) The condition is a "complete equi-join condition"; this is the best case scenario, the join is performed purely on a hash-based algorithm and no extra predicate is required.
b) The condition is a "partial equi-join conditiom", i.e. the condition contains some equi-join items, but also some non-equi-join items; in this case the join is performed on a hash-based algorithm (for the equi-join items) + an extra predicate (for the non-equi-join ones).
c) The join condition is a "complete non-equi-join-condition", i.e. there are no equi-join elements to build a hash-based solution, so the algorithm is performed based on a predicate which evaluates the whole condition. This is the worst-case scenario, since the HashJoin actually behaves as a kind of de-facto nested loop join.
Currently, since the condition nature is not evaluated in the computeSelfCost, cases a-b-c would have an equivalent cost; I think we should reflect somehow that: cost a < cost b < cost c

> EnumerableHashJoin should not use NLogN for costing
> ---------------------------------------------------
>
>                 Key: CALCITE-3657
>                 URL: https://issues.apache.org/jira/browse/CALCITE-3657
>             Project: Calcite
>          Issue Type: Bug
>          Components: core
>    Affects Versions: 1.21.0
>            Reporter: Vladimir Sitnikov
>            Priority: Major
>




--
This message was sent by Atlassian Jira
(v8.3.4#803005)