Abstract
The development of parallel Computational Fluid Dynamics (CFD) codes is a challenging task that entails efficient parallelization concepts and strategies in order to achieve good scalability values when running those codes on modern supercomputers with several thousands to millions of cores. In this paper, we present a hierarchical data structure for massive parallel computations that supports the coupling of a Navier–Stokes-based fluid flow code with the Boussinesq approximation in order to address complex thermal scenarios for energy-related assessments. The newly designed data structure is specifically designed with the idea of interactive data exploration and visualization during runtime of the simulation code; a major shortcoming of traditional high-performance computing (HPC) simulation codes. We further show and discuss speed-up values obtained on one of Germany’s top-ranked supercomputers with up to 140,000 processes and present simulation results for different engineering-based thermal problems.
Highlights
Due to ever-increasing advances in hardware, nowadays, high-performance computing (HPC) systems consist of several thousands to millions of cores, raising the necessity for sophisticated data structures and parallelization approaches in order to exploit the underlying computing power
The same data structure could be used to enable computational steering capabilities, where boundary conditions, fluid properties or geometry definitions could be changed during runtime, and an effect could be instantaneously observed in the simulation results
This paper presented a data structure capable of running efficiently on modern supercomputers, such as SuperMUC in Germany or Shaheen in the Kingdom of Saudi Arabia, and which is able to compute complex indoor airflow scenarios
Summary
Due to ever-increasing advances in hardware, nowadays, high-performance computing (HPC) systems consist of several thousands to millions of cores, raising the necessity for sophisticated data structures and parallelization approaches in order to exploit the underlying computing power. We have developed a hierarchical data structure that inherently supports distributed computing and combines state-of-the-art (numerical) solvers with efficient load-balancing techniques for fluid flow applications This approach could be successfully tested on different HPC systems, including one of Germany’s three top-ranked supercomputers, up to a total of 140,000 processes solving a problem with close to one trillion (1012 ) unknowns. The applications range from visualizing combustion problems (Yu et al [5]) to general procedures and software solutions (Fabian et al [6]) and extreme-scale scientific analysis (Bennett et al [7]) From reviewing these platforms and concepts, the authors decided to develop a new data structure, allowing one to solve some of the mentioned shortcomings and to provide a flexible environment for further model implementations. A dedicated section treats the code validation and verification according to standard benchmarks, such as the Rayleigh–Bénard convection, before looking into engineering-based examples generated from everyday engineering needs using complex temperature boundary conditions, as well as a complex geometric setup
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
