OpenSBLI: Automated code-generation for heterogeneous computing architectures applied to compressible fluid dynamics on structured grids
نویسندگان
چکیده
OpenSBLI is an open-source code-generation system for compressible fluid dynamics (CFD) on heterogeneous computing architectures. Written in Python, explicit high-order finite-difference solver structured curvilinear meshes. Shock-capturing performed by a choice of Weighted Essentially Non-Oscillatory (WENO) or Targeted (TENO) schemes. generates complete CFD the Oxford Parallel Structured (OPS) domain specific language. The OPS library embedded C code, enabling massively-parallel execution code variety high-performance-computing architectures, including GPUs. present paper presents base that has been completely rewritten from earlier proof concept (Jacobs et al, JoCS 18 (2017), 12-23), allowing shock capturing, coordinate transformations complex geometries, and wide range boundary conditions, solid walls with without heat transfer. A suite validation verification cases are presented, plus demonstration large-scale Direct Numerical Simulation (DNS) transitional Shockwave Boundary Layer Interaction (SBLI). shown to have good weak strong scaling multi-GPU clusters. We demonstrate languages suitable performing efficient simulations flows emerging
منابع مشابه
Telescoping Architectures: A Methodology for Evaluating Next-Generation Heterogeneous Computing
Architectural innovation has telescoped the HPC community from the commodity (Beowulf) cluster in a machine room, i.e., a multi-node system with Ethernet interconnect, to a commodity cluster on a chip, i.e., multicore CPU with an ondie interconnect. We project that this “telescoping architecture” will apply more broadly to heterogeneous computing, namely from heterogeneous clusters like Tianhe-...
متن کاملA Code Transformation Framework for Scientific Applications on Structured Grids
The combination of expert-tuned code expression and aggressive compiler optimizations is known to deliver the best achievable performance for modern multicore processors. The development and maintenance of these optimized code expressions is never trivial. Tedious and error-prone processes greatly decrease the code developer’s willingness to adopt manually-tuned optimizations. In this paper, we...
متن کاملPredictive Runtime Code Scheduling for Heterogeneous Architectures
Heterogeneous architectures are currently widespread. With the advent of easy-to-program general purpose GPUs, virtually every recent desktop computer is a heterogeneous system. Combining the CPU and the GPU brings great amounts of processing power. However, such architectures are often used in a restricted way for domain-specific applications like scientific applications and games, and they te...
متن کاملVirtual Environments for multiphysics code validation on Computing Grids
Multiphysics simulation is the core of future engineering design in the aerospace industry. For this to become a production reality, quantum leap breakthroughs are to be achieved, concerning in particular model coupling, error correlations, alert definitions, best usage practices, code verification and code validation. Because problems that are expected to be orders of magnitude larger than cur...
متن کاملExperiences with a computational fluid dynamics code on clouds, grids, and on-premise resources
Heterogeneity is often manifested in different forms even in environments that seem to be similar. In particular, clusters, grids, and infrastructure as a service (IaaS) clouds may appear straightforward to configure to be interchangeable. However, our experiences with mainstream parallel codes for solving complex computational fluid dynamics (CFD) problems demonstrate that secondary attributes...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computer Physics Communications
سال: 2021
ISSN: ['1879-2944', '0010-4655']
DOI: https://doi.org/10.1016/j.cpc.2021.108063