What is the status of the possibility of using java.nio.channels in James?
I have started to read about this package. It looks like it might have benefits for James or a project like James. In particular it seems to promise:
1) that blocked I/O operations on server Sockets can be interrupted gracefully
2) built in multi threading of connections
3) fancy buffering, somehow more efficient for large traffic
But I bet some people on this list have already studied it more than I, and formed valuable opinions. I'd like to learn those opinions.
java.nio has very little use unless you have an application where connections are kept open but data isn't being sent. HTTP, SMTP, and most common Internet protocols don't get much if any benefit from it. The biggest class of applications to benefit from java.nio is chat/IM where you have little data transfer and lots of open connections.
It would require a lot of rework to have the protocol handlers adjust to this, and this would only address the incoming connections, not the outgoing ones. We use standard libs for the outgoing ones, so we'd have to start at ground zero with those.
Basically, it's not worth it... just a technology that was overhyped (something related to Java was overhyped? impossible!) ;)
-- Serge Knystautas President Lokitech >> software . strategy . design >> http://www.lokitech.com p. 301.656.5501 e. [EMAIL PROTECTED]
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
