Yes. And I've done that scale of stuff on tiny machines. For stuff like
AberMUD5 it works far faster than splitting into processes. I even got
bored enough that with a generic (or its supplied) memory allocator it
locates data in memory in clusters to optimise paging ;)
Alan