Hacker News Reader

Comment by easterncalculus | Hacker News Reader

Seems very cool. I'm wondering what the next step would be for performance here. Trying python as suggested is quite slow so far.

  ~ # time python3 hello.py
  Hello World!
  real    0m 5.48s
  user    0m 4.11s
  sys     0m 1.24s

View on HN

There's a JIT, so running the same thing more than once makes it faster:

  ~ # time hello 
  Hello, world!
  real 0m 1.89s
  user 0m 0.67s
  sys 0m 1.14s
  ~ # time hello 
  Hello, world!
  real 0m 0.97s
  user 0m 0.36s
  sys 0m 0.55s

This would to me imply an obvious route of pre-warming the resumed image with your program. Or going all-in and natively compiling to wasm, but that's a much bigger lift.

I think it could be made much faster by emulating Linux system calls instead of running a real kernel, and translating x86 binaries to wasm before running (using something like cheerpx).

Reply on news.ycombinator.com