storing python function in distributed computer

finally I a little understand about tornadoweb asynchronous

the key of asynchronous response is

def new_messages(self, messages):
    cls = MessageMixin
    logging.info("Sending new message to %r listeners", len(cls.waiters))
    for callback in cls.waiters:
        try:
            callback(messages)
        except:
            logging.error("Error in waiter callback", exc_info=True)
    cls.waiters = []

so Is there any solution to store the callback function in datastore or solution if using several computer or process??

e.g : A requ开发者_开发知识库est updates to node1 B post data for A that handled by node2

how to node2 call node1 callback??

What?! You can only do callbacks with a message queue that has a data store. Look into something pika or Carrot. There's no way to "store" the callback but you can define a way to make other computers execute functions for you and have the results fed back to the original computer.