======================================================================== ftp.alumni.caltech.edu:/pub/ingber/MISC.DIR/parallel.txt Parallelizing ASA and PATHINT Project (PAPP) This file will keep a sparse running outline of this project being conducted under an award from the conducted under an award of Cray time from the Pittsburgh Supercomputing Center through the National Science Foundation (NSF). A more complete record of progress and selected correspondence will be released with the code. ======================================================================== $Id: parallel.txt,v 1.17 1994/07/25 11:04:10 ingber Exp ingber $ ======================================================================== The following is a list of people _currently_ active in PAPP. The reasons for this file are to start giving credit to people participating in PAPP and to start publicly documenting an outline of our development as part of the process of generating documented source code under the General Public License. Timothy Burns Utah Supercomputing Institute University of Utah Salt Lake City, UT 84112 PAPP Member: 13 Jun 94 Alan Cabrera Systems Development Sanwa Financial Products Co., L.P. 55 East 52 Street - 26 Floor New York, NY, USA 10055 PAPP Member: 17 Jun 94 Wolfram Gloger Poliklinik fuer Zahnerhaltung und Parodontologie Goethestr. 70 80336 Muenchen 2 Germany PAPP Member: 13 Jun 94 Lester Ingber Lester Ingber Research P.O. Box 857 McLean, VA 22101 PAPP Member: 27 May 94 Yushan Li Physics Department California Institute of Technology 106-38 Pasadena, CA 91125 PAPP Member: 16 Jun 94 Zhong Wu Dept. Mechanical and Aeronautical Engineering Clarkson University Potsdam, NY 13676 PAPP Member: 16 Jun 94 ======================================================================== @@ 5 Jun 94: First Posting Research Opportunities Parallelizing ASA and PATHINT I am looking for one to several people with experience parallelizing C code, e.g., on Crays, to work on parallelizing two specific algorithms: (a) Adaptive Simulated Annealing (ASA), and (b) an algorithm to calculate the time-development of multi-variable nonlinear Fokker-Planck-type systems, using a powerful non-Monte Carlo path integral algorithm (PATHINT). Some code and papers dealing with these algorithms can be obtained from ftp.alumni.caltech.edu [131.215.139.234] in the /pub/ingber directory. I am PI of an award of Cray time on an NSF Supercomputer, and have ported these codes successfully onto a C90. However, I am short of time to further optimize these codes, which is an essential requirement before doing production runs on C90 and T3D Crays. If necessary, I will do this work myself, but I would rather share the work, experience, and research with other interested people who also can expedite these projects. All code will remain under my copyright under the GNU General Public License (GPL), i.e., the least restrictive Library GPL. There are several immediate projects that are just waiting for detailed calculations, using codes which already run on SPARCstations, but need the power of supercomputers for production runs. All results of these studies will be published in peer-reviewed scientific journals, and only active participants on these projects will be co-authors on these papers. Examples of these projects include: neuroscience statistical mechanics of neocortical interactions (SMNI) EEG correlates of behavioral states short-term memory modeling realistic chaos + noise modeling financial applications 2- and 3-state term-structure security calculations testing of trading rules nonlinear modeling persistence of chaos in the presence of moderate noise If you are interested, please send me a short description of projects you have worked on, and how many hours/week you are prepared to commit to these projects for at least a period of 6-12 months. Lester ======================================================================== @@ 8 Jun 94: Reply to Many Respondents Thanks for replying to my posting. I am using the ASA_list mailing codes to protect the privacy of the respondents. There are several people with experience in parallelization and the applications I mentioned who also could commit to volunteer over 12 hrs/week for an extended period of time. I will contact these people separately, but here I do wish to thank you for your interest. If you did not explicitly tell me your expertise as it relates to these projects _and_ if you did not give a commitment of time over 1 day/wk for 6-12 months, then I did not consider you for the final group. There is plenty of work to do and our present team could always work with someone else with similar qualifications and time to commit. Some of you expressed interest in some of the projects mentioned and/or stated that you only had a little spare time. After our parallelization projects are completed and announced, please contact me again if you are interested in a specific project. As always, I am glad to answer specific questions on my ASA code or papers via e-mail correspondence. Questions on PATHINT likely should wait until it is released. Lester ======================================================================== @@ 8 Jul 94: After a reasonable trial period, the PAPP group will be consolidated to those people who actively participate in this experimental project. I expect this to be an ongoing process. Since communication on in-depth research is confined to e-mail and to postings on a "bulletin board" in my directory ~ingber/POSTINGS at PSC, members must use these channels if we are to communicate on various issues. New members who may join PAPP can quickly come up to the level of current participation by reading these postings and references contained therein to other documentation. We are looking at several schemes of parallelization, likely not doing full justice to all of them, including: BACS (lucy.ifi.unibas.ch:/bacs) PETSc (info.mcs.anl.gov:/pub/pdetools) PVM (http://pscinfo.psc.edu/general/software/packages/pvm/pvm.html) SISAL (sisal.llnl.gov:/pub/sisal) Lester ======================================================================== @@ 15 Jul 94: We have reached some agreement on how to proceed. It seems that using any formal language above PVM will take too much additional volunteer effort, not necessarily be the standard a year or two from now, and carry quite a bit of extra code (and possible bugs) that would be quite a strain to maintain on our volunteer effort across many platforms. ASA and PATHINT will continue to be coded in C, and using #ifdef's, etc., to use PVM on the Crays. In the spirit, but not in the detail, of existing higher languages and object oriented paradigms, we will write the PVM code at a level of communication at least one step higher than just the message passing interface to facilitate use of the code across some major platforms. The first project will be to get the present codes running and tested on the Crays, i.e., at the required ~100 MFLOP rate. The second project will be to crank out applications described in the proposal and in the original PAPP posting, resulting in some solid papers; these publications will be used to garner more resources for more projects. The third project will be to document the codes and to prepare them for the GPL release. Lester ========================================================================