Massive Parallelism was:(Re: oo hardware (was:...))
Jimmie Houchin
jhouchin at texoma.net
Sun Mar 23 03:45:46 UTC 2003
I know processes are cheap in Squeak and that doesn't necessarily
translate into 'processors' but how far is Squeak from such a scenario?
I also read the Erlang list and their basic philosophy seems to be to
create a process for most any activity. Lots and lots of processes.
Processes are cheap. Processes are how to model your solutions.
My apologies if I am not phrasing things correctly. I know there are
some here who are familiar with Erlang.
Erlang is designed for massive parallelism. Their processes are not
multi-processorable either. They run an new instance of the vm for each
processsor and do interprocess/vm communication via messages. Is Squeak
very far off from Erlang's massive parallelism? I have been considering
Erlang for web development. But I would prefer to stay with Squeak, but
I desire/need/want a very scalable solution.
This has been interesting.
But once again, I'll defer to those who know more than I on this. :)
Jimmie Houchin
Swan, Dean wrote:
[snip]
> 3) Regarding distributed OOP: I think an obvious solution
> is one CPU per object, with one or more shared high speed
> communication path(s) to send messages and return results.
>
> This could be reasonably explored on FPGA based hardware.
> Using a *really* simple CPU design, you could fit a lot of
> them on today's million gate FPGAs.
>
> I have often said that I would rather have a lot of slower
> CPUs than one really fast one. If the human brain can do
> all the wonderful things it does with a peak signal
> frequency of around 1 KHz, there must be something to this
> massive parallelism concept.
>
> As Alan mentioned, Fuchs was really on to something with
> his pixel processor idea. Too bad it hasn't caught on.
>
> Sadly, computer science has paid terribly little attention
> massive parallelism. Rumelhart and McClelland made some
> noise about this back in the mid-80's, but it's remained a
> niche field. Then there was the Transputer, Hillis's
> Connection Machine and others, but we've never really
> developed good tools to write parallel programs. The
> closest we have that is widely used is VHDL or Verilog.
>
> (This just caused an odd thought - How about a Smalltalk
> to digital logic compiler? Does that make any sense?
>
> After all, any CPU based architecture is always going to
> be sub-optimal compared to equivalent random logic.)
>
> -Dean
[snip]
More information about the Squeak-dev
mailing list
|