How can I use an aggregation strategy across nodes ?

Issue

• If I have a fuse cluster - and I am doing a splitter pattern and then writing to ActiveMQ. Can I create an aggregation strategy such that all nodes know when all the original file has been processed ?

• Can I split a message - put the parts on queues and then, once the message is totally split - aggregate messages based on some aspect of them ?

• Let's say that i have a message that looks like this

{code}
A:1:12341234
B:2:12341234123
C:1:123412341
D:1:1234123412
E:2:1234123412
{code}

Let's call each line a tuple of the form -- ID : secondary ID : value

• Can I split that by line - and send to a queue. Then, once ALL lines are processed, aggregate on the secondary ID ?