[
https://issues.apache.org/jira/browse/PHOENIX-7705?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18023187#comment-18023187
]
Kadir Ozdemir commented on PHOENIX-7705:
----------------------------------------
The row size function will work on the latest version or the latest version at
an SCN. Yes, we should have another function (RAW_ROW_SIZE()) to consider all
versions including delete markers. That should be another improvement on top of
this Jira.
> Support for a row size function
> -------------------------------
>
> Key: PHOENIX-7705
> URL: https://issues.apache.org/jira/browse/PHOENIX-7705
> Project: Phoenix
> Issue Type: Improvement
> Reporter: Kadir Ozdemir
> Assignee: Kadir Ozdemir
> Priority: Major
>
> Phoenix currently lacks a function to calculate the total size of a row,
> which is defined as the combined size of all HBase cells composing it. Such a
> function would be valuable for estimating a row's memory or storage footprint.
> For instance, this function could help configure the Phoenix client's
> mutation size limit for single-batch commits. Presently, assessing the size
> of a row is difficult because it involves not only column values but also the
> row key and other cell fields like column family, qualifier, timestamp, cell
> type, and MVCC sequence number.
> Another application is computing the storage footprint of a tenant within a
> multi-tenant table. This would aid in chargeback, monitoring, and service
> protection efforts.
> We propose introducing a function called ROW_SIZE to return the row's size.
> This function could then be used with SUM() to calculate the aggregate sum of
> row sizes (e.g., SUM(ROW_SIZE())).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
