Newsgroups: comp.parallel.mpi From: "David B. Serafini" Subject: Re: FFT's Organization: Lawrence Berkeley National Laboratory Date: Mon, 03 Aug 1998 19:24:04 -0700 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <35C670C4.446B@nersc.gov> > Aaron Focacci wrote: > >Does anyone know of where I could find an FFT subroutine implemented > >with MPI? There is an effort underway at the DOE NERSC center at Lawrence Berkeley National Lab to develop a portable, high performance parallel FFT implementation. I recently heard that the code is already considerably faster on a Cray than the Cray-supplied optimized FFT. I don't know much more. Look at for more info. with questions or comments. -- David B. Serafini Berkeley National Lab, MS 50B-2239 Bldg 50B, Room 3205 1 Cyclotron Road (510)495-2352 Berkeley, CA 94720