GPU Acceleration of Monte Carlo Simulation at the Cellular and DNA Levels
Geant4-DNA is an extension of the general purpose Geant4 Monte Carlo simulation toolkit. It can simulate particle-matter physical interactions down to very low energies in liquid water. The simulation in that energy scale needs enormous computing time since it simulates all physical interactions following a discrete approach. This work presents the implementation of the physics processes/models of the Geant4-DNA extension in GPU architecture. We observed impressive performance gain with the same physics accuracy as existing methods.
KeywordsSecondary Particle Global Memory Active Thread CUDA Kernel Neutral Hydrogen Atom
This work was supported by JSPS KAKENHI Grant 25246044, Japan-U.S. Cooperation in Research and Development in Science and Technology, and by the U.S. Department of Energy contract number DE-AC02-76SF00515. This work was partly supported by the Associated International Laboratory KEK (Japan)—CNRS (France)—CEA (France) “France-Japan Particle Physics Laboratory (FJPPL)”. The authors would like to thank NVIDIA for their generous support of this project and the CUDA Center of Excellence at Stanford University.
- 3.“The Geant4-DNA project”. http://geant4-dna.org (2014)
- 10.Henderson, N., Murakami, K. et al.: A CUDA Monte Carlo simulator for radiation therapy dosimetry based on Geant4. In: SNA + MC 2013—joint international conference on supercomputing in nuclear applications + Monte Carlo. doi:http://dx.doi.org/10.1051/snamc/201404204 (2014)
- 11.Murakami, K., Henderson, N. et al.: Geant4 based simulation of radiation dosimetry in CUDA. In: Nuclear science symposium and medical imaging conference (NSS/MIC). doi: 10.1109/NSSMIC.2013.6829452 (2013)