When an Agent wants to move to another Agent System it first asks the platform
for a proxy which implements the
AgentCommIFace (or derivates,
see Section
![[*]](/usr/share/latex2html/icons/crossref.png)
for details). Then it calls the
move
method of the proxy with its destination. The proxy handles the call straight
to the System Agent. The service then calls the
receiveAgent method
on the corresponding communication service of the target Agent System. This
can be done in the System Agent directly or through the output proxy. This depends
on the protocol which is implemented. If no exception occurs means that the
Agent was started correctly on the target System and can now be killed here.
Therefor calls the communication service the method
transferOK on its
SysAgentHelper. The helper wraps the call through to the platform
which terminates the Agent. See the process diagram
of moving an Agent.
