[
https://issues.apache.org/jira/browse/FLINK-5342?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15751031#comment-15751031
]
Fabian Hueske commented on FLINK-5342:
--------------------------------------
The cost model we are using for optimizing relational queries in flink-table is
very simple and not meant to give accurate cost estimates.
At the moment its main purpose is to push down filters and projections. We use
static input cardinalities and fixed selectivities for predicates because we do
not have any statistics about the input yet.
That's also the reason why we are not optimizing the order of joins.
Given the current situation, I don't think that we can use the cost model to
configure the parallelism of operators.
IMO, we should rather think about how we can offer more control to tune the
parallelism of operators.
> Setting the parallelism automatically for operators base on cost model
> ----------------------------------------------------------------------
>
> Key: FLINK-5342
> URL: https://issues.apache.org/jira/browse/FLINK-5342
> Project: Flink
> Issue Type: Improvement
> Components: Table API & SQL
> Reporter: godfrey he
>
> On Flink table API, a query will be translated to operators without
> parallelism. And user do not know even do not care the target operators
> translated from query. So it's better to set the parallelism automatically
> for each operator base on cost model.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
