I’m surprised at Elixir failing with 1M tasks. I’ve definitely done this sort of test with Erlang without an issue. Although I can’t really remember right now if I pushed it to 1M tasks.
https://www.erlang.org/doc/efficiency_guide/advanced.html#:~....
Maybe I’m cynical but this is what happens if you rely on chatGPT to give you programming answers without understanding the fundamentals first.
edit: oh apparently it’s got nothing to do with memory, 1M just happens to be higher than the default process limit for Erlang and thus Elixir.