[ 
https://issues.apache.org/jira/browse/CASSANDRA-21320?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Stefan Miklosovic updated CASSANDRA-21320:
------------------------------------------
          Fix Version/s: 6.0-alpha2
                         7.0
                             (was: 6.x)
                             (was: 7.x)
    Source Control Link: 
https://github.com/apache/cassandra/commit/bbff4f8d181b0a8d7440a85556d5ef0d7482b07f
             Resolution: Fixed
                 Status: Resolved  (was: Ready to Commit)

> Track rows written per write request as histogram metric (similar to rows 
> read)
> -------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-21320
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21320
>             Project: Apache Cassandra
>          Issue Type: Improvement
>          Components: Observability/Metrics
>            Reporter: Piotr Walczak
>            Assignee: Stefan Miklosovic
>            Priority: Normal
>             Fix For: 6.0-alpha2, 7.0
>
>          Time Spent: 50m
>  Remaining Estimate: 0h
>
> Hello team,
> I am planning to make necessary changes to cover below issue:
> Problem:
> Currently, Cassandra provides histogram metrics for reads to show rows 
> scanned per read request (e.g., rows_per_read  - using liveScannedHistogram). 
> However, there is no equivalent metric on the write side that tracks rows 
> written per write request at the table level.
> Solution
> Add a new histogram metric rowsMutatedPerWriteHistogram that:
> 1. Tracks the total number of rows mutated in each write request at the table 
> level
> 2. Is recorded once per write request 
> 3. Provides equivalent visibility into write patterns as rows_per_read does 
> for read patterns
> Use Case
> Operations teams can now analyze write patterns alongside read patterns:
>  
> Code changes proposal: 
> [https://github.com/apache/cassandra/compare/trunk...pwalczak:cassandra:pwalczak/CASSANDRA-21320/trunk?expand=1]
> 1. What's the typical size of write requests (in terms of rows)?
> 2. How does write request size distribution compare across tables?
> 3. Identify unusually large batch writes that may impact performance
> 4. Monitor trends in write request patterns over time



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to