[jira] [Created] (CASSANDRA-21321) Add RowsRead and RowsMutated counters to TableMetrics for accurate per-table row throughput tracking

["Piotr Walczak (Jira)" <ji...@apache.org>]

Piotr Walczak created CASSANDRA-21321:
-----------------------------------------

             Summary: Add RowsRead and RowsMutated counters to TableMetrics for accurate per-table row throughput tracking
                 Key: CASSANDRA-21321
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21321
             Project: Apache Cassandra
          Issue Type: Improvement
            Reporter: Piotr Walczak


Cassandra already exposes {{LiveScannedHistogram}} (rows returned per read) and {{RowsMutatedPerWriteHistogram}} (rows touched per write - https://issues.apache.org/jira/browse/CASSANDRA-21320) at the table level. These histograms are valuable for understanding the *distribution* of rows per operation, but they are *insufficient for measuring total row throughput* over a time window — which is essential for capacity planning and predictions.

Specifically:
 * A histogram records _how many rows a single operation touched_ (e.g., "this query returned 47 rows").
 * To derive total rows from a histogram you would need to multiply bucket midpoints by counts — an approximation that becomes increasingly inaccurate for skewed distributions.
 * Histograms also do not give you a simple delta between two scrape timestamps, making windowed rate calculations fragile.

A *monotonically-increasing counter* solves all of these problems: scrape the value at {{t1}} and {{{}t2{}}}, subtract, divide by the interval — exact rows/sec with no approximation.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org