So it's basically amb, but with using threading instead of backtracking. So probably faster than the classic amb implementation, esp. on multiprocessor architec

So it's basically amb, but with using threading instead of backtracking. So probably faster than the classic amb implementation, esp. on multiprocessor architectures, but also much more resource intensive.