db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Harshvardhan Gupta (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DERBY-6942) Utilise additional statistics for selectivity estimates.
Date Wed, 28 Jun 2017 19:51:00 GMT

    [ https://issues.apache.org/jira/browse/DERBY-6942?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16067124#comment-16067124
] 

Harshvardhan Gupta commented on DERBY-6942:
-------------------------------------------

In particular, when estimating cost for a particular conglomerate, it would be good if we
start looking at other conglomerates' statistics for cardinality estimates. For example during
table scan, we can start adjusting cardinality estimates using the statistics for index.

Note that cost and cardinality are two different concepts, cost will always be determined
by the number and size of pages / rows that a particular access path has to read to execute
its work. Table scan's cost will remain unaffected here in this case by changing thecardinality
estimates. (However, index scan's cost and cardinality estimates will be co-related in our
particular example)

> Utilise additional statistics for selectivity estimates.
> --------------------------------------------------------
>
>                 Key: DERBY-6942
>                 URL: https://issues.apache.org/jira/browse/DERBY-6942
>             Project: Derby
>          Issue Type: Sub-task
>          Components: SQL
>            Reporter: Harshvardhan Gupta
>            Assignee: Harshvardhan Gupta
>            Priority: Minor
>




--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message