Spawning processes seam to be much more efficient in Erlang than starting new threads using the OS (i.e. by using Java or C). Erlang spawns thousands and millions of processes in a short period of time.

I know that they are different since Erlang do per process GC and processes in Erlang are implemented the same way on all platforms which isn't the case with OS threads.

But I don't fully understand technically why Erlang processes are so much more efficient and has much smaller memory footprint per process. Both the OS and Erlang VM has to do scheduling, context switches and keep track of the values in the registers and so on...

Simply why isn't OS threads implemented in the same way as processes in Erlang? Does it have to support something more? and why does it need a bigger memory footprint?

Technically why is processes in Erlang more efficient than OS threads? And why can't threads in the OS be implemented and managed in the same efficient way?