<div dir="ltr">Hi,<br><br>We are going to write a C-node for communication with our Erlang node. <br>And we are going to use the erl_interface library.<br><br>I need information on how many sockets are used in nodes communication? <br>One socket for a node? One socket for a message? E.t.c.<br>I'd like to know this for understanding how differentiates the "classic" model of <br>work with sockets (http server, e.t.c.) and model of work when writing C-node.<br><br>"Classic" for me (server side, brief) is:<br><br>Â Â Â Â socket<br>Â Â Â Â Â Â |<br>Â Â Â Â Â bind<br>Â Â Â Â Â Â |<br>Â Â Â +-> listen(select)<br>Â Â Â | Â Â |<br>Â Â Â ------+ --'new thread'<br>Â Â Â Â Â Â Â Â Â Â |<br>Â Â Â Â Â Â Â Â Â accept<br>Â Â Â Â Â Â Â Â Â Â |<br>Â Â Â Â Â Â Â Â recv/send<br>Â Â Â Â Â Â Â Â Â Â |<br>Â Â Â Â Â Â Â close (accept)<br>Â Â Â Â Â Â Â <br>Is it relevant for C-node? One accept per one client/connection? <br>Or one accept per one erlang message?<br></div>