The future of plasma simulations

Plasmas constitute paradigmatic examples of complex physical systems, involving nonlinear, multiscale processes far from thermodynamic equilibrium. HPC and virtual prototyping has today an increasingly crucial role in explaining and predicting observations as well as in designing more effective plasma systems and manufacturing processes. This provides novel pathways to reducing the risks, costs, and time often associated with the design and construction of new devices, and large experimental facilities, such as fusion and accelerator devices. Plasma physics has been a key driver for HPC ever since the 1960s, and continuing into the exascale era. The Max Planck Computing and Data Facility (MPCDF, part of the proposing consortium) in Germany and the National Energy Research Supercomputer Center (NERSC) in the U.S. both started out (in 1960 and 1974, respectively) with a clear focus on plasma applications. Moreover, when the Roadrunner supercomputer at the Los Alamos National Laboratory became the first supercomputer to break the sustained petascale barrier, one of the lighthouse codes was a plasma simulation code called VPIC, demonstrating the capabilities in simulating laser plasma interaction.

Like for the petascale, lighthouse plasma simulation codes will be among the very first to break the exaflop barrier. Plasma-PEPSC aims at enabling four lighthouse plasma simulation codes (BIT, GENE, PIConGPU, Vlasiator) from different plasma physics domains (plasma-material interfaces, fusion, accelerator physics, space physics) and important scientific drivers to exploit exascale supercomputers.

Plasma-PEPSC aims at providing efficient computational tools helping address several Grand Challenges in plasma physics, employing high-fidelity descriptions based on kinetic models: optimizing magnetic confinement fusion devices, developing new accelerator technologies, and predicting space weather. In this context, we focus our efforts on four European flagship plasma simulation codes with a large user base, bringing together a complementary and interdisciplinary team of world-leading domain scientists from plasma physics, applied mathematics, and computer science (including HPC, computational science, software engineering, and data analytics).

To achieve these ambitious scientific goals, we want to maximize the performance achievable by our four codes and enable them on the current pre-exascale and upcoming exascale supercomputers by algorithmic improvements (automatic load- balancing, compression, and resilience), performance optimisation for highly heterogeneous systems (accelerators and heterogeneous memories) and high-throughput online data analysis. Together with enabling the tackling of scientific grand challenges on exascale supercomputers, Plasma-PEPSC co-designs the four plasma simulation codes with the systems developed within EPI by providing feedback to the design of European processors, accelerators, and pilots and investigating the plasma simulation code changes to exploit these systems with maximum efficiency.