[CASSANDRA-19676] Stream processing for StorageProxy::updateCoordinatorWriteLatencyTableMetric

• Type: Improvement
• Status: Resolved
• Priority: Normal
• Resolution: Fixed
• Fix Version/s: 5.0-rc1, 5.1
• Component/s: Legacy/Core
• Change Category: Performance
• Complexity: Low Hanging Fruit
• Platform: All

Source Control Link

https://github.com/apache/cassandra/commit/1b4f898c32a8f882767f44d69b38ff85d85090ba

Test and Documentation Plan

ci

• Assignee: Sam Lightfoot
• Reporter: Sam Lightfoot

Authors

Sam Lightfoot

Reviewers

Stefan Miklosovic

• Created: 02/Jun/24 16:25
• Updated: 06/Jun/24 15:54
• Resolved: 06/Jun/24 10:54

Description

On profiling a write-heavy workload (90% writes) using easy-cass-stress, it became very clear StorageProxy::updateCoordinatorWriteLatencyTableMetric was a hot path that ~15% of the CPU cycles of ModificationStatement::executeWithoutCondition were taken up by (see attached async-profiler image).

We should convert this stream to a simple for loop, as has been discussed recently on the mail list.

easy-cass-stress command:

$ bin/easy-cass-stress run KeyValue -n 10m --maxwlat 10 -r 0.1 --rate 20000 --compaction twcs

 

Issue Links

• is cloned by: CASSANDRA-19679 Stream processing for SimpleRestriction::bindAndGetClusteringElements
• is related to: CASSANDRA-19673 Investigate stream pipelines in hot paths
• links to: GitHub Pull Request #3344