Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
2001
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*December
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 2001

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Two CPU AMD PC and Mathematica

  • To: mathgroup at smc.vnet.net
  • Subject: [mg29657] Re: Two CPU AMD PC and Mathematica
  • From: Mike Yukish <may106 at psu.edu>
  • Date: Mon, 2 Jul 2001 02:20:19 -0400 (EDT)
  • Organization: Penn State University, Center for Academic Computing
  • References: <9guok5$nv9$1@smc.vnet.net> <9h1ajj$rhu$1@smc.vnet.net> <9h3vq9$48n$1@smc.vnet.net> <9hc8d2$40r$1@smc.vnet.net> <9hetk5$6k8$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Orestis Vantzos wrote:

> The single most common Mathematica operation I use is Map (and variations of
> it, such as Table)...I don't see why it couldn't be made to run
> concurrently. Divide the elements of the list to be operated on and spread
> among available processors, or something like that. I figure there would be
> an impressive speed up of my programs, if this could be done...
> Orestis

Hi,

I have the parallel processing toolbox. It has a ParallelMap[ ] function which
does exactly what you want. If you use it, though, you'll find it probably more
computationally efficient to split the list to be mapped on into sublists, and
send those sublists to the processors. With N elements in a list and M
processors, if you use the ParallelMap[ ] function you get the comm overhead of
N calls, but if you split the list and send out sublists you get the comm
overhead of M calls.





  • Prev by Date: RotateShape
  • Next by Date: Re: .m file to C or Fortran source code
  • Previous by thread: RotateShape
  • Next by thread: Re: .m file to C or Fortran source code