One more thought on 1sided

Tony Kimball (alk@think.com)
Wed, 22 Nov 95 14:48:20 EST

If MPI on any of these machines actually handles progress by interrupting
the application's CPU, I would like to make a note of it.

I have not read existing MPI implementations, but there is certainly a
whole slew of MP libraries for the CM5 which rely upon network interrupts,
and a few compilers as well. Unless one is doing a blocking synchronous
receive, it is the (overwhlemingly) preferred approach. Any MPI
implementation for the CM5 which does not talk to the hardware directly is
almost certainly handling progress by interrupting the CPU. I cannot speak
to other platforms.