A general idea of the kind of applications that run on Roadrunner can be seen in the SC'08 Initial report. The application described there is a neutron-transport model. The kind of the communication pattern seen in this application is described as a wavefront. For the Roadrunner it was ported from a pure MPI implementation. On Roadrunner the Opterons are used for communication and the compute intensive part is ported on to the PowerXCell.
http://www.c3.lanl.gov/pal/publications/papers/barker08:SC08_Roadrunner.pdf On Fri, Apr 5, 2013 at 9:18 AM, Vincent Diepeveen <d...@xs4all.nl> wrote: > Note that as for nuke explosions i have no idea how those look like - > maybe someone more knowledgeable wants to comment on that. > > As a total layman there i would suspect that it's important where the > protons/neutrons/whatever-tons/supertiny-tons are located. I'd be > modelling that naively using > matrixcalculations. > > So that would mean the only low level library you need is a > matrixcalculation and some relative simple functions - with the > matrixcalculations > eating 99% of all system time on that massive supercomputer out of > all calculations done on it. > > In such case one would need surprisingly little very well optimized > code to make optimal usage out of such massive supercomputer. > > Any other 'secret' batchjob i'd be running on a different > supercomputer. If there is no need to run a massive vector oriented > workload type matrixcalculation - > one just shouldn't run on such type of supercomputer i feel. NASA > still had that 10240 socket supercomputer back then if i remember well, > to give one example... > > > > On Apr 4, 2013, at 5:39 PM, Max R. Dechantsreiter wrote: > > > Vincent, > > > >> It doesn't matter whether you code for blue gene, cuda or phi - from > >> a software viewpoint it's all vector type coding you've got to do. > >> the price of 1 coder is total peanuts compared to > >> the price of those supercomputers. So specialistic written software > >> is what you need anyway. > > > > Simply porting an application shouldn't take much effort, > > as long as vendor-specific libraries aren't involved. > > However, special and sometimes intensive efforts are often > > required to achieve good performance (high utilization). > > > > The unfortunate fact is that funds are more easily spent > > on hardware than on the human resources needed to utilize > > them effectively. > > > > Max > > --- > > http://www.linkedin.in/in/benchmarking > > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf >
_______________________________________________ Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
