Hi, I would like to ask some questions and expose some ideas of mine.
1) How many people might need a simulator for more than one application?
2) How many people might use simulators as means of debugging their
application?
3) How many people might use simulators to evaluate application performance
and behavior?
I found myself in need of a simulator to debug the algorithms for
interaction between two applications. So I posted a question about how to
hammer to applications in TOSSIM and later I had an idea. So, how hard would
it be to translate nesC to C++? Given C++ code for components, it would be
possible to build and use a NS-like simulator, no? So I kept thinking about
this translation process:
There could be two ways to bind components via interface: by generating
code for the bindings (static) or by passing objects (which would be
instances of components) as parameters to other objects in execution time
(dynamic). Also, statically bound interfaces could be inspected many times
in order to remove indirection levels. Some low-level components would need
to be implemented as C++ code integrated to the simulator if efficiency
during simulation were important, otherwise the simulator could emulate the
nodes right away.
I have to put more thought on the other ideas I have.
Regards,
Bernardo
--
"The truth shall set you free"
_______________________________________________
Tinyos-help mailing list
[email protected]
https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
