perf-review

Profile and benchmark jlsm components to identify performance bottlenecks.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/nathannorthcutt/jlsm --skill perf-review-nathannorthcutt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: perf-review
Source: https://github.com/nathannorthcutt/jlsm/tree/main/.claude/skills/perf-review
Command: npx skills add https://github.com/nathannorthcutt/jlsm --skill perf-review-nathannorthcutt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a rigorous, repeatable process to identify, measure, and reason about performance bottlenecks in the jlsm LSM-Tree library, enabling data-driven optimizations.

Core Features & Use Cases

  • Measure CPU, memory, and IO behavior across MemTable, SSTable, and compaction paths.
  • Profile and instrument hot paths, allocations, and GC pressure to guide targeted optimizations.
  • Generate actionable findings with location, hypothesis, evidence, and concrete fixes; track progress across commits and branches.

Quick Start

Start a performance session on a target module using perf-review to collect measurements and generate findings.

Frequently Asked Questions about perf-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I identify performance bottlenecks in an LSM-Tree library?

Identify performance bottlenecks in an LSM-Tree library by running systematic benchmarks and profiling CPU, memory, and IO behavior across MemTable, SSTable, and compaction paths to generate evidence-based optimization recommendations.

What is the best way to profile memory allocations and GC pressure in Java?

Profile memory allocations and GC pressure in Java by instrumenting hot paths to measure resource usage, generating actionable findings that include location, hypothesis, evidence, and concrete fixes to guide targeted optimizations.

Can I use benchmark results to track performance improvements across commits?

Track performance improvements across commits by using benchmark sessions to generate and record actionable findings, allowing you to quantify resource behavior changes and compare optimizations over time.

How do I measure CPU and IO behavior during SSTable compaction?

Measure CPU and IO behavior during SSTable compaction by profiling the compaction paths within the LSM-Tree library, systematically collecting metrics to pinpoint specific resource bottlenecks.

Does this performance analysis approach require any external dependencies?

This performance analysis approach requires no external dependencies, providing a rigorous and repeatable process using only systematic benchmarks and profiling to drive data-driven optimizations.