Starting to lay down the design of a system (single CPU, single application) that will basically have to do the 
following:
  communicate with "outside" world via radio data link
  communicate with "internal" pieces via RS-232
  manage messaging, data transformation, etc between the internal    pieces & outside world
  internal pieces subject to change (new hardware for instance)

One of the problems/issues I am having is that after digging thru the docs for the past 7 - 10 days; I am stiil not 
convinced that I understand the relationships between partitions, processes and threads. Seems as if partitions have 
processes which in turn have threads (or some combination thereof). For example, I want the external I/O to have its own
 time slice (partition) and address space (process) and within it have thread(s) sitting on the input ports and moving 
message data to/from another area (partition/process/thread(s)) which does the work. 
Does this make sense from a QNX 6.4 standpoint or am I missing something (or maybe a lot of somethings).
Thanks for any light you may be able to shed