[Maxima] is maxima suitable for multiprocessor computations?
fateman at cs.berkeley.edu
Thu Nov 15 19:03:17 CST 2007
I think that designs for efficient parallel algorithms on distributed
networks of loosely-coupled machines will be qualitatively different from
algorithms on shared-memory machines, and any attempt to cover over these
differences by an abstraction layer is likely to make a shambles of the
efficiency aspect of the algorithm. Since the only aspect of interest in
doing things in parallel is the efficiency, what do we have?
Multiple Maximas communicating with each other via sockets versus one
Maxima that (say) can do a tree-search in memory with multiple processors...
I'm not a fan of PVM..
> -----Original Message-----
> From: maxima-bounces at math.utexas.edu
> [mailto:maxima-bounces at math.utexas.edu] On Behalf Of Larry Prevett
> Sent: Thursday, November 15, 2007 4:49 PM
> To: maxima at math.utexas.edu
> Subject: Re: [Maxima] is maxima suitable for multiprocessor
> For those interested, Paul Wang did some papers regarding
> parallel processing
> and at least some of the work involved maxima.
> from links on this page:
> L. Prevett
> Maxima mailing list
> Maxima at math.utexas.edu
More information about the Maxima