A Python extension for the massively parallel multiphysics simulation framework waLBerla

Abstract

We present a Python extension to the massively parallel HPC simulation toolkit waLBerla. waLBerla is a framework for stencil based algorithms operating on block-structured grids, with the main application field being fluid simulations in complex geometries using the lattice Boltzmann method. Careful performance engineering results in excellent node performance and good scalability to over 400,000 cores. To increase the usability and flexibility of the framework, a Python interface was developed. Python extensions are used at all stages of the simulation pipeline: They simplify and automate scenario setup, evaluation, and plotting. We show how our Python interface outperforms the existing text-file-based configuration mechanism, providing features like automatic nondimensionalization of physical quantities and handling of complex parameter dependencies. Furthermore, Python is used to process and evaluate results while the simulation is running, leading to smaller output files and the possibility to adjust parameters dependent on the current simulation state. C++ data structures are exported such that a seamless interfacing to other numerical Python libraries is possible. The expressive power of Python and the performance of C++ make development of efficient code with low time effort possible.

waLBerla: A block-structured high-performance framework for multiphysics simulations

“…While the GUI's main goal is to assist the developer during debugging through visualization of internal data structures, it is not targeted at the user of the final simulation application. Instead, waLBerla provides a powerful Python interface for simulation setup, computational steering, as well as result evaluation and visualization [128].…”

lbmpy: Automatic code generation for efficient parallel lattice Boltzmann methods

“…This fully parallel data structure enables adaptive grid refinement and dynamic load balancing between MPI processes [4]. WAL-BERLA has Python bindings [6] that allow for simple distributed simulations with lbmpy generated kernels directly from Python on a uniform grid. For advanced use cases, like grid refinement, the user has to switch to C++ as the driving language.…”

Code generation approaches for parallel geometric multigrid solvers

“…Therefore, in the last years, various approaches to automate the process to derive a simulation code from a mathematical model by code generation have been explored. Our group conducted research in several projects on both external and embedded domain-specific HPC languages in Python [5], Scala [30], AnyDSL [52], C++ and CommonLisp [26].…”