The Immersed Boundary-Lattice Boltzmann Method Parallel Model for Fluid-Structure Interaction on Heterogeneous Platforms

<table class="table-group" id="tab3"><tr><td><table class="table"><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr class="thead"><td class="align_left">Domain size</td><td class="align_center">CPUMFLUPS</td><td class="align_center">OpenMPMFLUPS</td><td class="align_center">GPUMFLUPS</td><td class="align_center">CPU-GPUMFLUPS</td></tr><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr><td class="align_left">62 5× 89 (55625)</td><td class="align_center">0.874</td><td class="align_center">19.841</td><td class="align_center">193.750</td><td class="align_center">193.835</td></tr><tr><td class="align_left">750 × 107 (80250)</td><td class="align_center">0.953</td><td class="align_center">23.902</td><td class="align_center">240.037</td><td class="align_center">240.849</td></tr><tr><td class="align_left">900 × 128 (115200)</td><td class="align_center">0.958</td><td class="align_center">24.057</td><td class="align_center">294.730</td><td class="align_center">295.036</td></tr><tr><td class="align_left">1080 × 154 (166320)</td><td class="align_center">0.956</td><td class="align_center">24.793</td><td class="align_center">321.114</td><td class="align_center">322.597</td></tr><tr class="table-tr"><td colspan="5"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Double-precision performance of CPU-GPU heterogeneous platform.</div>

Mathematical Problems in Engineering

tab3

Table 3

Table 3: The Immersed Boundary-Lattice Boltzmann Method Parallel Model for Fluid-Structure Interaction on Heterogeneous Platforms 