high performance computing on graphics processing units: hgpu.org

The presented thesis establishes simulations on modern massively parallel computing hardware to investigate relativistic laser-driven plasmas. The latter are of special interest as they may provide a compact source for energetic ion beams. Computer simulations provide valuable insight into ultrafast plasma processes, evolving in the ultrahigh intensity (l_0 >> 1018 W/cm^2) focus of the ultrashort (tau_0 = 30 – 500 fs) laser pulses driving the interaction. Such simulations require high numerical resolution and full geometric treatment for reliable predictions, which can only be addressed with high-performance computing. The open source particle-in-cell code PIConGPU, which is developed in the framework of this thesis, answers these demands, providing speed and scalability to run on the world’s largest supercomputers. PIConGPU is designed with a modular and extensible implementation, allowing to compute on current and upcoming hardware from a single code base. Furthermore, challenges arising for generated data rates, reaching 1 PByte per simulation, are resolved with scalable data reduction techniques and novel workflows, such as interactive simulations. Numerical studies are performed on two novel targets for laser-proton acceleration with near-critical and mass-limited properties. A micrometer-scale spherical target is explored with realistic temporal laser contrast, providing an interpretation for experimental results collected at the PW-class laser system PHELIX (tau_0 = 500 fs pulse length). In this study, 3D modeling with the GPU supercomputer Titan enabled the identification of pre-expansion to near-critical target conditions, which uncovers a regime of volumetric laser-electron interaction generating a highly directed proton beam. Furthermore, a novel cryogenic hydrogen jet target is researched in close collaboration to experiments at the laser system DRACO (τ0 = 30 fs). This target system provides a unique setup for the isolated investigation of multi-species effects and their influence on the generated ion energy distribution. A novel analytical model provides a link between characteristic modulations in the ion energy spectra and ensemble properties of the microscopic electron distribution. In view of a potential experimental realization, parametric scans are performed confirming the feasibility of the proposed setup.