Asynchronous RPC Over the Named-pipe Protocol

If you use named pipes ( ncacn_np)as your transport protocol you should avoid allowing a large number of idle pending calls on the server. With named-pipes, each client waiting for a reply will have a pending named-pipe-read on the server, each of which requires a certain amount of kernel memory.

For example, you would not want to use a notification call for new e-mail with the named-pipe transport, because such a call would remain pending even when clients are idle, and kernel memory could be depleted. Note that this is not a problem with the other connection-oriented protocols, such as ncacn_ip_tcp.