A phenomenon I’ve noticed: If a proposed parallel extension to a language doesn’t provide speedups to programs that are greater than the speedups that users would get by just rewriting those programs in a different language, the extension is probably doomed.
Good enough that the gains you get from rewriting your CPU-bound OCaml code in e.g. C++ are usually better than the gains you would get from parallelizing the OCaml.
-
-
Yes, but then you would be using C++, and losing out on the benefits of OCaml. Adding multi-core support to OCaml both 1) allows you to gain from multi-core, and 2) allows you to continue using OCaml. Plus, multi-core is tied up with effects, so you gain that benefit too on top
-
your assumption seems to be that there might not be a benefit beyond speed to write in a language
- 10 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.