MathGroup Archive 2000

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

Search the Archive

Re: Parallel programming.

  • To: mathgroup at smc.vnet.net
  • Subject: [mg22969] Re: Parallel programming.
  • From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
  • Date: Sat, 8 Apr 2000 14:44:41 -0400 (EDT)
  • Organization: Universitaet Leipzig
  • References: <8cha2f$975@smc.vnet.net> <8ck2um$gf0@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Hi,

> 
> It sounds to me like your quad SPARC might work pretty well for this
> but I cannot make any guarantees.  Alternatively, a good fast new
> PC costing about $2,000 might actually give you better performance than
> the quad Sparc.  Furthermore, if you were really into it,
> you could get parallel Mathematica running on a cluster of PC boxes using
> 100BT interconnect.

A 100 BM ethernet connection is about 10-50 times slower than the 
link on the same machine (depending in the net collisions). 
But the effect depends strongly on
the ratio parallel/serial code. With a program that has 99 %
parallel code the conection does not realy  matter. With a program
that does only 50 % parallel you can be dramatical slower.

Regards
  Jens


  • Prev by Date: Re: uninstalling mathreader
  • Next by Date: Re: Command to get a notebook's directory?
  • Previous by thread: Re: Parallel programming.
  • Next by thread: Mathematica NDSolve max number of grid points error