question on prototypying tools for concurrent systems (answers)

[Seung Mo Cho]

Hello.

I've got a number of replies. Thank you.
However, I should say that none of them seems adequate for my purpose.

Most people recommended simulation tools, which generally accompany
symbolic analysis tools, such as SPIN, TwoTowers, mCRL, CADP, etc.
These tools usually enable interactive simulation or random testing.

But, what I want is a tool that enables 'builing' a program that
would play the similar role of human in interactive simulation :
collecting data
by inspecting the execution, determining which transitions to execute
next (based on the collected data), and firing (a) step(s) of
transition(s) of a model, etc.
Therefore, what I need is a kind of scripting language for programming
schedulers.

Pawel let me know that programming languages based on concurrency
theories, such nomadic Pict, may be good for modeling purpose,
but, he's not heard of implementations with the funcitionalities I
mentioned.
I agree with him.

One possibility might be MAUDE, a rewrite logic tool.
I am aware that it has the funtionality of reflection,
which in turn enables programming scheduler. But, it looks somewhat
too difficult to learn and use, for non-experts of rewrite logic, at
least at the
current stage of development...

Thank you for reading (and replying) my mail.

Seung Mo Cho.

Seung Mo Cho wrote:

>Hello.
>
>
>I am looking for a tool which allows prototying and controlled
>testing of concurrent systems. The functionalities I need are
>as follows. (Maybe the second and the third are a little bit difficult.)
>
>
>- modeling systems with EFSM (extended finite state machines)
>or similar formalisms allowing interleaving concurrency model.
>
>- controling the execution of the model with programable scheduler.
>(i.e. allowing specification of execution strategy.)
>
>- collecting information during the simulation, and feeding them
>back to the execution (possibly through the scheduler)
>
>
>As far as I know, the IOA simulator from MIT seems the closest one.
>But it is not so much developed yet, and therefore many features
>are missing, currently. (For example, backtracking the simulation
>is not supported, and the language for describing the scheduler is
>not expressive enough.)
>
>If there are some useful ones, please let me know.
>
>Thank you.
>
>Seung Mo Cho.
>
>
>-- Seung Mo Cho, (Ph. D.) Reseach Associate, Social Communication
>Laboratory NTT Communication Science Laboratories Telephone :
>+81-774-93-5273 e-mail : sm...@cslab.kecl.ntt.co.jp msn : seu...@msn.com
>
>
>
>

--
Seung Mo Cho, (Ph. D.)
Reseach Associate,
Social Communication Laboratory
NTT Communication Science Laboratories

Telephone : +81-774-93-5273
e-mail : sm...@cslab.kecl.ntt.co.jp
msn : seu...@msn.com