UK BOINC Team Forum

When a gravitational wave is observed, answering the question "What exactly produced this?" is crucial to advancing science. Inferring physical properties of even the simplest observed gravitational wave source-black hole binaries-requires catalogs of numerical relativity gravitational waveforms spanning all seven dimensions of intrinsic parameter space (i.e., mass ratio, plus the three spin vector components of each black hole). Due to the requirement that virtually all numerical relativity simulations to date be run on supercomputers, all such catalogs combined sample this parameter space to only about 3 points per dimension.

These tiny catalogs have been sufficient for noisy gravitational wave observations to date, as the noise acts to obscure the relatively small effects of misaligned spins, but they will not be good enough moving forward.

My BlackHoles@Home project aims to reduce the cost of numerical relativity black hole and neutron star binary simulations by ~100x, through adoption of numerical grids that fully exploit near-symmetries in these systems. With this cost savings, black hole binary merger simulations can be performed entirely on a consumer-grade desktop (or laptop) computer.

BlackHoles@Home is destined to run on the BOINC infrastructure (alongside SETI@Home), enabling anyone with a computer to contribute to construction of the largest numerical relativity gravitational wave catalogs ever produced.