Performance characteristics of a parallel treecode

dc.creatorValdarnini, R.
dc.date2002-12-11
dc.date.accessioned2026-07-07T02:06:50Z
dc.date.available2026-07-07T02:06:50Z
dc.descriptionI describe here the performances of a parallel treecode with individual particle timesteps. The code is based on the Barnes-Hut algorithm and runs cosmological N-body simulations on parallel machines with a distributed memory architecture using the MPI message passing library. For a configuration with a constant number of particles per processor the scalability of the code has been tested up to P=32 processors. The average CPU time per processor necessary for solving the gravitational interactions is within $\sim 10 %$ of that expected from the ideal scaling relation. The load balancing efficiency is high ($\simgt90%$) if the processor domains are determined every large timestep according to a weighting scheme which takes into account the total particle computational load within the timestep.
dc.description10 pages, 3 figures in "Computational astrophysics in Italy: methods and tools'', Bologna, July 4-5, 2002, Ed R. Capuzzo Dolcetta
dc.identifierhttps://arxiv.org/abs/astro-ph/0212256
dc.identifierhttp://arxiv.org/abs/astro-ph/0212256
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/5498
dc.subjectAstrophysics
dc.titlePerformance characteristics of a parallel treecode
dc.typetext

Files

Collections