I've been reading Gary Sabot's THE PARALATION MODEL: Architecture-Independent
Parallel Programming (MIT Press, 1988), and I'm considering using his model
in a large Common Lisp project. For those who are not familiar with this
work, here's a short blurb from the jacket:
"THE PARALATION MODEL introduces a way of programming parallel computers
that is easy to use for general problem solving, and will work for many
different parallel computer architectures with any number of processors,
from one to billions. The book includes working LISP source code for a
mini-compiler, along with many programming examples."
For those who ARE familiar with the work, I have a few questions:
Accessibility: I'd like to get a copy of the serial implementation which
will run in Macintosh Allegro Common Lisp. The book came with an order form,
but I'd like to know if the information and price are still current, and
if there is some way (maybe free for academics ;->) to get it by FTP.
I also want to know about the accessibility of the connection machine
implementation; I do have access to a CM, and this is one of the reasons that
I'm considering the Paralation model.
Real-life testimonials: I'd like to hear from anyone (other than the system's
designers, who are probably a bit biased) who has used the paralation model
either successfully or unsuccessfully. Particularly helpful would be comments
about how easy it has been to re-write existing systems within the model, and
about the likelyhood of getting significant speedups with the CM compiler and
a seemingly MIMD application.
Thanks for any information you can provide. I will summarize email responses
to the net if there is interest.
-Lee Spector (·······@cs.umd.edu)
P.S. I couldn't find any parallelism-related news groups to send this to;
if you know a more appropriate forum for this discussion I'd appreciate hearing
about it...