[comp.parallel] Parallelism and OOPSs

nobody@acsu.buffalo.edu (10/06/89)

In article <477@schaefer.MATH.WISC.EDU> wayne@schaefer.MATH.WISC.EDU (Rick Wayne) writes:
>
>since objects decompose so nicely into independent entities (at least 
               ^^^^^^^^^^----> Isn't it better :
since objects are independent entities (at least 
				?
>abstractly), wouldn't this provide an excellent hook for introducing 
>parallelism?  could be loosely or tightly coupled; message passing would
>either be memory-to-memory transfer or network sends.
>
>this sounds so obvious that i'm sure SOMEBODY must be doing it
what about actors?
>                          comments?
	I think this could be a real interesting topic for further
discussions (also for comp.parallel).

	Hopefull

	Giancarlo

charmi@cs.buffalo.edu


[ I certainly second the motion that this group has a real interest
in the subject.

	Steve
]

jxw@RODS.IUS.CS.CMU.EDU (John Willis) (10/10/89)

Philips Research is actively building a parallel processor system
structured around objects.  The machine / system is called DOOM,
while the language is called POOL.  A 100 processor system is
already reported operational.  Several non-trivial applications
have been written for the machine by other organizations.  If you
are interested in the potential of objects for structuring
parallel computation, find out more about DOOM.  There was a good
survey article in IEEE Computer or IEEE Micro about 2 years ago
in the issue on International Computing.  Also see the 1987 and
1989 proceedings of the PARLE conference (available from Springer
-Verlag).