Parsing Millions of URLs per Second
watch on portal.gitnation.org ↗https://portal.gitnation.org/contents/parsing-millions-of-urls-per-secondsummary
This covers how Node.js URL parsing was sped up by about 3× from Node 16 to Node 20 by reducing instruction counts and vectorizing sub-algorithms, allowing Node.js to match and surpass rivals like Bun and Deno while still supporting the WHATWG URL standard. It describes concrete optimizations applied to both the C++ and JavaScript sides of the Node.js implementation to achieve millions of URLs parsed per second.
AI-written summary · about
related talks
11mNode.js Performance Optimizations
24mYou Did What? Understanding Node.js Optimizations
11mRevolutionizing JavaScript Performance
34mNode.js Can Be as Fast as C++
53mMastering Node.js Performance
23mDecomposition of the Main Thread in Node.js to increase throughput
31mThe State of Node.js Core
25mFaster than C? Parsing Node.js Streams!