Computer Science
Scientific paper
Aug 2006
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2006iaujd...6e..33h&link_type=abstract
Neutron Stars and Black Holes in Star Clusters, 26th meeting of the IAU, Joint Discussion 6, 17-18 August 2006, Prague, Czech Re
Computer Science
Scientific paper
Simulations of dense stellar systems, such as globular clusters and galactic nuclei, have never yet been very realistic. Simplifying assumptions, such as used in Fokker-Planck codes or gas models, have allowed us to model large particle numbers at the expense of a loss of detail in local many-body interactions and the imposition of global symmetry constraints. Conversely, direct N-body integration, while far more accurate, has labored under a lack of computer speed needed to model a million stars. The good news is that we will soon be approaching effective computer speeds in the Petaflops range, which will allow us to model the gravitational million-body problem with full realism, at least on the level of point particles. Adding equally realistic stellar evolution and hydrodynamics will be no problem as far as the necessary computer speed is concerned. When the hardware bottleneck will thus be removed, the software bottleneck for realistic cluster simulations will soon become painfully obvious. This is the bad news. While some serious uncertainties remain in the science needed to improve the software, currently the main bottleneck is neither science nor computer speed, but rather a sufficiently robust implementation of already available knowledge.
No associations
LandOfFree
Direct N-body Simulations does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Direct N-body Simulations, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Direct N-body Simulations will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1783833