David W. Pfitzner, John K. Salmon
Cosmological N-body simulations on parallel computers produce large datasets - about five hundred Megabytes at a single output time, or tens of Gigabytes over the course of a simulation. These large datasets require further analysis before they can be compared to astronomical observations. We have implemented two methods for performing halo finding, a key part of the knowledge discovery process, on parallel machines. One of these is a parallel implementation of the friends of friends (FOF) algorithm, widely used in the field of N-body cosmology. The new isodensity (ID) method has been developed to overcome some of the shortcomings of FOF. Both have been implemented on a variety of computer systems, and successfully used to extract halos from simulations with up to 2563 (or about 16.8 million) particles, which are among the largest N-body cosmology simulations in existence.