Re: Parallel Toolkit
- To: mathgroup at smc.vnet.net
- Subject: [mg23310] Re: Parallel Toolkit
- From: Jens-Peer Kuska <kuska at informatik.uni-leipzig.de>
- Date: Mon, 1 May 2000 18:00:11 -0400 (EDT)
- Organization: Universitaet Leipzig
- References: <8eilmh$kdu@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
Hi, the Parallel Toolkit works fine for Master/Slave models. But it depends on your task and your model to setup other topologies. I have tryed a ring and a torus topologie. How ever if you make a pure simulation of some stupid numerics a C-program possible with MPI will always faster. The toolkit has it's power in parallel symbolic computations. Hope that helps Jens msc wrote: > > Greetings, > > Does anyone have experience using the newly released Parallel Toolkit, > particularly for simuation work? > > Thanks, > > Mark