Effective Performance Testing to your Server with Autocannon
summary
Autocannon is used to benchmark Node.js servers and diagnose a synchronous CPU‑heavy hash blocking the event loop, then compared to an asynchronous worker‑thread version with measurable gains: 1.5s 99th percentile → 1.4s, average latency ~800ms → ~700ms and 12 rps → 14 rps (peak 20). Demonstrates CLI flags (-c, -d, -w), difference between HTTP pipelining and concurrent connections, scripted request flows that POST, capture an id and POST again, and recommends using production‑like datasets.
AI-written summary · about
chapters (5)
- 0:00Start
- 6:43The Important Questions We Need To Ask
- 10:57HTTP Request Pipelines VS Concurrent Connections
- 21:28AutoCannon Actually Use Worker Threads
- 22:24Basic Example
related talks
53mMastering Node.js Performance
46mNodeJS Scalibility Tips
26mObserving Node.js: Using Metrics to Improve your Application Performance
01h 12mA New Way to Profile Async Activity in Node.js- 53mMastering Node.js Performance
01h 04mNode.js performance analytics and best practices
40mPerf Stories
36mPerformance Optimization 101