[erlang-questions] gproc scalability, shared ets, links and using terminate/2

[Max Lapshin]

Yes, perhaps problem is because client process dies, sends DOWN
message and tracker
is busy with cleaning data from ets.

If client will send {demonitor_me, Pid} message, so tracker will know
that ets is cleaned and it will behave much faster.

I'll try that, because problems, that Ulf is telling, may be very serious.
100K of messages is not an issue. Erlang VM can deal with them in 2
seconds. Printing of them is a problem.