11 ms·
no (in both directions). lots of np/exp problems paralize well and you can be in NC and parallelize really inefficiently (e.g. you can get a 10x speedup, but yo
by adgjlsfhk1 4mo ago
no (in both directions). lots of np/exp problems paralize well and you can be in NC and parallelize really inefficiently (e.g. you can get a 10x speedup, but you need 1000000x the hardware). the better framing is that NC is the class of efficient algorithms that can be sped up near arbitrarily by parallelization
- osti 4mo agoHmm your last sentence seems to exactly agree that it's a class of algos that parallelize well? What does sped up arbitrarily mean? It's still polynomial speed up right?
- chowells 4mo agoIt's a difference of degree. People expect something that "parallelizes well" to show near 1-to-1 speedup. Double the hardware, double the speed. This is "you can always speed it up, but the hardware requirements can increase at any polynomial rate".
- osti 4mo agoAh got it. Reread previous comment and that makes sense.
- dragontamer 4mo agoYeah it's more of "on a hypothetical infinitely parallel computer, you'll get a big speedup'. Which is still useful if you can prove a problem is in NC. It's just not quite as strong as people make it out to be.