[
https://issues.apache.org/jira/browse/PHOENIX-1453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14259913#comment-14259913
]
James Taylor commented on PHOENIX-1453:
---------------------------------------
I was thinking total byte count flattened across each guidepost per region. We
could alternatively just keep a single total row count that we can use with the
total byte count we're already keeping (so we can calculate an avg byte count
per row), as this covers the use case where we'd use this now.
We don't have a lot of uses for row count yet - we could hold off on making
this change until we have more. More immediate ones are PHOENIX-1312 (as we're
currently never using stats for other column families), and PHOENIX-1565 (just
filed to more accurately adjust byte count value when a split occurs).
[~maryannxue] - can you think of more use cases for having a row count in
addition to a byte count?
> Collect row counts per region in stats table
> --------------------------------------------
>
> Key: PHOENIX-1453
> URL: https://issues.apache.org/jira/browse/PHOENIX-1453
> Project: Phoenix
> Issue Type: Sub-task
> Reporter: James Taylor
> Assignee: ramkrishna.s.vasudevan
> Attachments: Phoenix-1453.patch, Phoenix-1453_1.patch,
> Phoenix-1453_10.patch, Phoenix-1453_2.patch, Phoenix-1453_3.patch,
> Phoenix-1453_7.patch, Phoenix-1453_8.patch
>
>
> We currently collect guideposts per equal chunk, but we should also capture
> row counts. Should we have a parallel array with the guideposts that count
> rows per guidepost, or is it enough to have a per region count?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
