Hello,
Does anyone know of a parallel lisp? Common lisp flavor is preferred.
And, can anyone provide me info. as to how to call C functions in AKCL?
Thanks in advance for any help.
Wei Jen Yeh ···@cs.purdue.edu
Department of Computer Science
Purdue University
West Lafayette, Indiana
--
Wei Jen Yeh ···@cs.purdue.edu
Department of Computer Science
Purdue University
West Lafayette, Indiana
In article <·····@medusa.cs.purdue.edu> ···@cs.purdue.EDU (Wei Jen Yeh) writes:
> Does anyone know of a parallel lisp? Common lisp flavor is preferred....
>
I've been looking for a good //el implementation for while.
As far as I know there isn't one (I say that in the hope that someone
will prove me wrong!)
The nearest I've seen (heard of) was a version of Betz's xlisp
running on Helios on a transputer array.
Robert Hamilton
Dept of Electrical Eng.
University of Edinburgh.
Email: ····@uk.ac.ed.ee
In article <····@castle.ed.ac.uk>, ······@castle.ed.ac.uk (R Hamilton) writes:
> In article <·····@medusa.cs.purdue.edu> ···@cs.purdue.EDU (Wei Jen Yeh) writes:
> > Does anyone know of a parallel lisp? Common lisp flavor is preferred....
> >
> I've been looking for a good //el implementation for while.
> As far as I know there isn't one (I say that in the hope that someone
> will prove me wrong!)
> The nearest I've seen (heard of) was a version of Betz's xlisp
> running on Helios on a transputer array.
Parallel Lisp is a research area (albeit a fairly active one). By that
I mean that, while it's an area of considerable interest and work,
it's by no means certain that you're going to find something that will
suit your purposes. However, we can do better than transputer-specific
systems.
Most people working on parallel Lisp address specifically parallel
machines, but there are often implementations which time-share on
serial machines. There are various possible mechanisms for
parallelising Lisp, though, and they differ greatly in approach and
applicability. I'm presuming that the original poster wanted to run a
parallel Lisp on a serial machine.
Thinking Machines have a simulator for *Lisp (as used on the
Connection Machine -- see Danny Hillis' book) which is available for
anonymous ftp. Other interesting contenders are QLisp (Stanford) and
Multilisp (MIT), although last time I looked neither of those were
widely available for serial machines. It would be hard to imagine
three systems which took more different approaches to parallelising
Lisp!
An early implementation of EuLisp, a Lisp dialect whose definition is
currently being worked on by a CEC-sponsored group, incorporates a
thread mechanism, and is freely available (to get a copy, try mailing
······@maths.bath.ac.uk). It's not entirely Common Lisp-like; whether
this is a good or bad thing is a matter of debate.
For Rob Hamilton, my own contribution is in the building next door (a
Linda-based parallel Lisp running on the Edinburgh Concurrent
Supercomputer)!
-- Paul.
--
Paul Dourish, Rank Xerox EuroPARC, Cambridge, UK <·······@europarc.xerox.com>
"Ain't they got no barbers where you come from, boy?"
It is very hard to by-pass this oportunity to talk about EuLISP.
EuLISP was designed to provide parallel processing, and there are
primitive for just that. The FEEL implementation runs on our shared
memory Stardent, and also runs pseudo-parallel on SUN-OS. We believe
that porting FEEL is not very hard, so moving to other shared memory
machines should be OK.
Yes this is a research program and so things change, but the current
state etc can be determined by sending e-mail to
······@uk.ac.bath.maths
Code is free (FEEL = Free and Eventually Eu Lisp)
==John ff
> Does anyone know of a parallel Lisp?
The one that comes to my mind was the one described in Danny Hillis'
dissertation, "The Connection Machine". They have a list variant
called a "xector", in which each element is processed simultaneously
by a processor on the CM. I don't remember if it is very CL compliant.
In article <·····················@bilver.uucp> ········@bilver.uucp (Dave Andrews) writes:
> Does anyone know of a parallel Lisp?
Lucid has a parallel lisp for the sequent called CLiP (Common Lisp in Parallel)
which we have here at NMSU. Somewhat buggy, but does have some nice features
too.
--
tom eskridge
computing research laboratory, new mexico state university
········@nmsu.edu (505) 646-6247
In article <·····················@bilver.uucp> ········@bilver.uucp (Dave Andrews) writes:
>The one that comes to my mind was the one described in Danny Hillis'
>dissertation, "The Connection Machine". They have a list variant
>called a "xector", in which each element is processed simultaneously
>by a processor on the CM. I don't remember if it is very CL compliant.
You are describing CM Lisp. It is completely CL compliant, because it is
simply an orthogonal extension to CL.
Unfortunately, it was never implemented fully, as far as I know. CM Lisp
was a very ambitious design, because it abstracted a good deal away from
the Connection Machine hardware, and expected many Common Lisp primitives
to deal with xectors directly.
Instead, what we implemented was *Lisp, which is much closer to the CM
architecture, and requires programmers to use a separate set of constructs
for dealing with "pvars" (parallel variables) and parallel operations.
Functions that accept pvars are generally named <name>!! and map the CL
function <name> over the elements of the pvars (e.g. +!!); global and
control constructs are generally named *<name>, where <name> is the
corresponding CL operator (e.g. *IF, *SUM).
Gary Sabot developed Paralation Lisp as part of his dissertation. I
believe his book "The Paralation Model" includes a disk with a simulator on
it.
--
Barry Margolin, Thinking Machines Corp.
······@think.com
{uunet,harvard}!think!barmar
In article <····················@Think.COM>, ······@think.com (Barry Margolin) writes:
> Gary Sabot developed Paralation Lisp as part of his dissertation. I
> believe his book "The Paralation Model" includes a disk with a simulator on
> it.
I have that book and there was no disc in it.
--
The purpose of advertising is to destroy the freedom of the market.