Hey Max, Nice to hear from you and I'm glad you're making progress!
> for the last couple of years I have been toying with hpx. I have now > reached a state where for some simple, but practically relevant, > problems I get performance that is comparable with openmp on a single > node. My main motivation for turning to hpx is to have have multi-node > parallel code. I have started to work on that based on the > documentation and a few other sources, but I have not gotten very far. > So I am looking for an hpx-guru or at least someone with experience > getting some code to run on multiple nodes. > > The bit of code I am looking to parallelize initially is very simple. > The core algorithm has about 200 lines (without setup, configuration > etc.) and the problem is embarrassingly parallel. So for someone who > knows what they are doing, this should be a piece of cake. Still, > there are a couple of interesting aspects (e.g. using the new > partitioned vector) and I have two other, slightly more complicated, > bits that have been prepared for parallelization. > > So this is what I am looking for: > > - - Someone to extend the existing parallelization to multiple nodes > using hpx > - - Compile and run the existing benchmark codes to quantify the scalabili > ty Please count me in. Regards Hartmut --------------- http://boost-spirit.com http://stellar.cct.lsu.edu > > > In return I offer: > > - - Co-authorship on a paper describing the parallelization and > performance improvement. I plan to submit it to Computers & Geosciences > (http://www.journals.elsevier.com/computers-and-geosciences/). I can > write most of the paper, so this would be a relatively minimal effort > for a peer-reviewed publication > > - - Access to the University 255 node cluster for the duration of the > project > > - - Continuing collaboration if that is of interest. > > > If this sound interesting to you or you have any questions, just send > me an email and we can discuss this further. > > Regards > > Max > > - -- > This message, its contents and related information will be analyzed by > NSA, GCHQ and most likely others. No information in this message can > therefore be regarded as confidential. If you want to communicate with > me confidentially, please contact me about encrypted communication. > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2.0.22 (GNU/Linux) > > iQEcBAEBAgAGBQJVUIbYAAoJEPGnJUm5Ir26KO8IAMVK9PzSJ9CVmpryzwCdOeN/ > nGZTTFTNCRD7590si7ShUzuWxivs6hOLq+vK7HfmEPbqomotgeaAkQtQNN3O4I9p > Pexd69eBfV13Z5MvmwIJAw4lSJlXMUqvGdSuCxxK26tpPJ/oSTpegAJX9kLucgNg > DndZm1osWM6riJzhsFI1I7NRLlDpm6NY1b/TU9hgkPP6HZtRnLbfaq2PNAuccwwP > V7Jwl+7Mm5QupHm2J1WRvB8HDyNRdMn1wR54pRp4cd5wU8IXEOQKWXMC7S4yxvKX > dsicxKbrba9tk4LFWt468HDBdC75VAUKpF9lYnzGNFMLP8eivX2nJGHkCDXrJlk= > =DvO4 > -----END PGP SIGNATURE----- > _______________________________________________ > hpx-users mailing list > [email protected] > https://mail.cct.lsu.edu/mailman/listinfo/hpx-users _______________________________________________ hpx-users mailing list [email protected] https://mail.cct.lsu.edu/mailman/listinfo/hpx-users
