I built a 2x faster lexer, then discovered I/O was the real bottleneck(modulovalue.com)

5 ポイント投稿者 modulovalue6 か月前1 コメント

https://modulovalue.com/blog/syscall-overhead-tar-gz-io-performance/

1 コメント

jll296 か月前

Nice story; that's exactly why people always warn from the eval called "premature optimization without measurement".