Hacker News new | ask | show | jobs
by aheilbut 4657 days ago
Imagine the type of machine learning we could do on 384 low power 1GHz cores.
1 comments

Not everything parallelizes nicely. Given enough processors, serial computation eventually becomes the bottleneck:

http://en.wikipedia.org/wiki/Amdahl's_law

Many machine learning methods parallelize extremely nicely.