Skip to content

Conversation

@Lakshith7748
Copy link

Description

This PR enhances the default SQL query histogram configuration by introducing a wider range of bucket boundaries to improve real-world latency visibility.
Previously, the upper bound was limited to 10 ms, compressing slower queries into a single bucket.
Now, when no custom buckets are provided, a comprehensive default set (0.001 s to 60 s) is applied to capture micro to long-running query latencies effectively.

Key change:

Updated NewHistogram() in metrics.go to use a broader default bucket range.

Added validation for monotonically increasing bucket boundaries.

Benefits:

Improved accuracy for query latency analysis.

Better differentiation between fast, slow, and critical queries.

More meaningful P95/P99 latency metrics for production workloads.

Breaking Changes:

None
Existing code providing custom bucket values remains fully compatible.

Additional Information:

Changes are limited to metrics.go under the metrics package.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant