Not all index scans are equal: How we cut query latency by over 99%
Datadog | The Monitor blog

Not all index scans are equal: How we cut query latency by over 99%


Summary

Datadog engineers optimized a slow PostgreSQL query by replacing an inefficient composite index scan with a targeted index that better matches the query's predicates. This change, which addressed a mismatch in B-tree column order, slashed average query latency from 300 ms to just 38 μs. Additionally, the article highlights how Datadog Database Monitoring can now automatically detect and suggest remedies for similar suboptimal index scans.
Read the Original Article

This article originally appeared on Datadog | The Monitor blog.

Read Full Article on Original Site

Popular from Datadog | The Monitor blog

1
Datadog LLM Observability natively supports OpenTelemetry GenAI Semantic Conventions
2
Introducing Bits AI Dev Agent for Code Security
Introducing Bits AI Dev Agent for Code Security

Datadog | The Monitor blog Mar 26, 2026 77 views

3
Monitoring MongoDB performance metrics (MMAP)
Monitoring MongoDB performance metrics (MMAP)

Datadog | The Monitor blog May 25, 2016 70 views

4
Understand session replays faster with AI summaries and smart chapters
Understand session replays faster with AI summaries and smart chapters

Datadog | The Monitor blog Apr 2, 2026 69 views