4 ms·

Note also that these numbers include startup time, which is noticeably slower for JVM compared to node. If you're not restarting the server on every request (li

Note also that these numbers include startup time, which is noticeably slower for JVM compared to node. If you're not restarting the server on every request (like the "serverless" nonsense does), a second more or less is not that important. I'd like to see sustained throughput comparison between those two.

igouy 5y ago
For those tiny tiny programs 1/10th of a second JVM startup —- “Wtf kind of benchmark counts the jvm startup time?” https://benchmarksgame-team.pages.debian.net/benchmarksgame/sometimes-people-just-make-up-stuff.html https://benchmarksgame-team.pages.debian.net/benchmarksgame/...