Performance of a Code Migration for the Simulation of Supersonic Ejector Flow to SMP, MIC, and GPU Using OpenMP, OpenMP+LEO, and OpenACC Directives

<table class="table-group" id="tab6"><tr><td><table class="table"><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr class="thead"><td align="left">GPU </td><td align="center"> Computing time </td><td align="center"> Versus CPU serial vectorized without HT </td><td align="center"> Versus CPU (HT) OpenMP vectorized </td><td align="center"> Versus MIC </td></tr><tr><td class="thead-hr" colspan="5"><hr/></td></tr><tr><td align="left"> C2070 </td><td align="center"> 03 h 31 m 47 s (1.0X) </td><td align="center"> 9.86X </td><td align="center"> 1.6X </td><td align="center"> 4.5X </td></tr><tr class="table-tr"><td colspan="5"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

Comparison of the speed-up factors taking as reference the GPU.

Scientific Programming

tab6

Table 6

Table 6: Performance of a Code Migration for the Simulation of Supersonic Ejector Flow to SMP, MIC, and GPU Using OpenMP, OpenMP+LEO, and OpenACC Directives 

Table 6 | Performance of a Code Migration for the Simulation of Supersonic Ejector Flow to SMP, MIC, and GPU Using OpenMP, OpenMP+LEO, and OpenACC Directives 