is maxima suitable for multiprocessor computations?
Subject: is maxima suitable for multiprocessor computations?
From: Stavros Macrakis
Date: Sun, 25 Nov 2007 13:43:18 -0500
On Nov 22, 2007 11:06 AM, Robert Dodier <robert.dodier at gmail.com> wrote:
> Along these lines, I came across this blurb about MapReduce which
> appears to be a system for automatically parallelizing programs.
>
Not really. It is a programming abstraction and an efficient implementation
on loosely-coupled networks of machines. The name makes it sounds as though
it is something like xreduce(reducef,map(mapf,l)), but it isn't -- it
actually operates on mappings (not lists), and is considerably more general.
As for where the abstraction came from, the paper says it "is inspired by
the map and reduce primitives present in Lisp and many other functional
languages." However, it is considerably more general, and it seems to have
much more of the flavor of SETL than of Lisp (even if the authors of the
paper haven't heard of SETL!).
-s