Profiling & Optimization

Table of Contents

• Overview
• When to Use
• Quick Start
• Reference Guides
• Best Practices

Overview

Profile code execution to identify performance bottlenecks and optimize critical paths using data-driven approaches.

When to Use

• Performance optimization
• Identifying CPU bottlenecks
• Optimizing hot paths
• Investigating slow requests
• Reducing latency
• Improving throughput

Quick Start

Minimal working example:

typescript
1import { performance, PerformanceObserver } from "perf_hooks";
2
3class Profiler {
4  private marks = new Map<string, number>();
5
6  mark(name: string): void {
7    this.marks.set(name, performance.now());
8  }
9
10  measure(name: string, startMark: string): number {
11    const start = this.marks.get(startMark);
12    if (!start) throw new Error(`Mark ${startMark} not found`);
13
14    const duration = performance.now() - start;
15    console.log(`${name}: ${duration.toFixed(2)}ms`);
16
17    return duration;
18  }
19
20  async profile<T>(name: string, fn: () => Promise<T>): Promise<T> {
21    const start = performance.now();
22
23    try {
24      return await fn();
25    } finally {
26// ... (see reference guides for full implementation)

Reference Guides

Detailed implementations in the references/ directory:

Best Practices

✅ DO

• Profile before optimizing
• Focus on hot paths
• Measure impact of changes
• Use production-like data
• Consider memory vs speed tradeoffs
• Document optimization rationale

❌ DON'T

• Optimize without profiling
• Ignore readability for minor gains
• Skip benchmarking
• Optimize cold paths
• Make changes without measurement