HiGPUs: Hermite's N-body integrator running on GPUs

Discussion topics for individual codes
Post Reply
User avatar
Guardian of the Codes
Posts: 1141
Joined: Wed Aug 04, 2004 4:18 pm
Location: Washington, DC

HiGPUs: Hermite's N-body integrator running on Graphic Processing Units

Post by owlice » Sat Jul 14, 2012 8:20 am

HiGPUs: Hermite's N-body integrator running on Graphic Processing Units

Abstract: HiGPUs is an implementation of the numerical integration of the classical, gravitational, N-body problem, based on a 6th order Hermite’s integration scheme with block time steps, with a direct evaluation of the particle-particle forces. The main innovation of this code is its full parallelization, exploiting both OpenMP and MPI in the use of the multicore Central Processing Units as well as either Compute Unified Device Architecture (CUDA) or OpenCL for the hosted Graphic Processing Units. We tested both performance and accuracy of the code using up to 256 GPUs in the supercomputer IBM iDataPlex DX360M3 Linux Infiniband Cluster provided by the italian supercomputing consortium CINECA, for values of N ≤ 8 millions. We were able to follow the evolution of a system of 8 million bodies for few crossing times, task previously unreached by direct summation codes.

HiGPUs is also available as part of the <a href="http://ascl.net/1107.007">AMUSE</a> project.

Credit: Spera, M.; Capuzzo Dolcetta, R.; Punzo, D.

Site: http://astrowww.phys.uniroma1.it/dolcet ... iGPUs.html

Bibcode: 2012ascl.soft07002S

ID: ascl:1207.002
Last edited by Ada Coda on Sun Jul 18, 2021 3:43 am, edited 1 time in total.
Reason: Updated code entry.
Code-seeking owl at your service

Post Reply