CyclesPhi
We have modified the kernel of the Blender Cycles rendering engine and then extended its capabilities to support the HPC environment. We call this version the CyclesPhi and it supports following technologies:
- OpenMP
- MPI
- Intel® Xeon Phi™
Benchmarks v1
I used the scenes from this page: https://code.blender.org/2016/02/new-cycles-benchmark/
Computed devices:
- GTX970 ( Win )
- KNL ( Intel Xeon Phi 7250, https://www.hlrn.de/ )
- KNC ( Intel Xeon Phi 7120P, https://docs.it4i.cz/salomon/hardware-overview/ )
- OMP24 ( 2 x Intel Xeon E5-2680v3, 2.5 GHz, 12 cores )
- OMP24+MIC2 ( 2 x Intel Xeon E5-2680v3, 2.5 GHz, 12 cores + 2 x Intel Xeon Phi 7120P )
The time is without preprocessing of the scene ( BVH, … ). Blender 2.77 was compiled with ICC.
Benchmarks v2
I used the scenes from this page: https://code.blender.org/2016/02/new-cycles-benchmark/
Computed devices:
- IT4Innovations
- Salomon
- 2x Intel® Xeon™ E5-2680v3 (HASWELLS)
- NVIDIA® GeForce® GTX TITAN X
- Workstation (Windows 10)
- NVIDIA® GeForce® GTX 970
- Salomon
- Marconi-A2 CINECA
- Intel® Xeon Phi™ 7250 (KNL)
Blender 2.79 was compiled with GCC.
Run CyclesPhi
Copiyng
Blender uses the GNU General Public License, which describes the rights to distribute or change the code. For Intel libraries please read intel_license.
Links