Subject: Re: C466: Redistribution strategies for portable parallel FFT: A case study Resent-Date: Wed, 01 Nov 2000 12:48:52 -0500 Resent-From: Geoffrey Fox Resent-To: Geoffrey Fox Date: Wed, 1 Nov 2000 11:44:00 -0600 (CST) From: Anshu Dubey To: Geoffrey Fox On Sat, 30 Sep 2000, Geoffrey Fox wrote: > I have gotten referee reports on your paper and we woulkd be happy to publish > it with two changes. Namely both referees noted that some of the the figures are rather difficult > to read. I wonder if you could not replot something like #Processors * Execution Time > which is rouighly efficiency which would allow more "resolution" in a given size figure. > You could also make figures bigger. Also one referee would like the concepts of distributed > FFT (as opposed to transpose) to be explained for general reader > > Please send me a manuscript addressing above points to me electronically (PDF or PS) > and it can be published without further ado) > Thank you for your interest in Concurrency:Practice and Experience > > 8506444587 > We would like to thank you very much for and your input about the paper. Attached with this mail is the postscript file of the paper, with the following modifications as per the referee's suggestions : 1. We have included a brief description of the distributed FFT's in the "Introduction" section, highlighting their differences from the transpose based FFT's. We have also included the appropriate reference for more detailed description. 2. We have changed the scale of the unreadable figures to log-log scale, which has given them a great deal of clarity. Please let me know if these modification do not meet the requirements, or if you would like any other changes to the paper. Regards Anshu Dubey Astronomy and Astrophysics University of Chicago 5640 S. Ellis Avenue Chicago, IL 60637 (773) 834 2999 --------------------------------------------------------------------- Name: paper.ps paper.ps Type: Postscript Document (APPLICATION/PostScript) Encoding: BASE64