Yes it is quite SEDA-ish and also like ActiveMQ in-VM queues. However, while there is a queue similarity I think probably we should implement our own code at this point. If we find this becomes complex or tricky we could look at other codebases. But maybe I'm oversimplifying this aspect in my mind.

Paul

On 10/19/05, ant elder <[EMAIL PROTECTED]> wrote:


On 10/18/05, Paul Fremantle < [EMAIL PROTECTED]> wrote:
We do need a well-defined API call for a mediation to "issue" a message directly into the "bus"

This is interesting, so thats like a new transport listener for local messages. There'd be a thread switch so the program putting the new message doesn't have to wait for it to be processed. There'd be a pool of threads to process the local messages, and I quess the messages would sit in a queue waiting for an available thread.

All sounds very SEDAish, and it just like a Queue in a messaging system. Are we going to write that from scratch or use something like ActiveMQ in-VM queues?

   ...ant


Reply via email to