Performance and Scalability of Unstructured Mesh CFD Workflow on Emerging Architectures
نویسندگان
چکیده
The performance and scalability of an unstructured mesh based CFD workflow on Intel Xeon Phi based system is discussed.
منابع مشابه
Understanding the Parallel Scalability of An Implicit Unstructured Mesh CFD Code
In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examining different programming models. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presenting primarily experimental...
متن کاملParallel unstructured mesh CFD codes: a role for recursive clustering techniques in mesh decomposition
In principle, unstructured mesh CFD codes can be parallelised using a mesh decomposition approach similar to structured mesh codes. However, for unstructured codes the mesh structure is problem dependent and algorithms for automatically decomposing the mesh onto the processors are required. An algorithm based upon a recursive clustering technique, for decomposing meshes into an arbitrary number...
متن کاملOptimizing CAD and Mesh Generation Workflow for SeisSol
SeisSol is a simulation software for seismic wave propagation and earthquake scenarios. It solves the fully elastic wave equations in heterogeneous media. Incorporating dynamic rupture simulation it performs complex multiphysics earthquake simulations. To account for complicated geometries SeisSol uses a fully unstructured tetrahedral mesh. Recent publications [1], [2] have shown that SeisSol i...
متن کاملDevelopment of a Pressure-Based CFD Solver for All-Speed Flows on Arbitrary Polygonal Meshes in a Rule-Based Framework
Over the last decade significant new technologies have been developed which allow the scientist/engineer to design reliable programs for simulating the complex physics involved in turbulent flow combustion devices. In the area of numerical algorithm development, a significant step forward has come with the maturation of unstructured grid methods, which provide a host of benefits over traditiona...
متن کاملScalability of parallel finite element algorithms on multi-core platforms
The speedup of element-by-element FEM algorithms depends not only on peak processor performance but also on access time to shared mesh data. Eliminating memory boundness would significantly speed up unstructured mesh computations on hybrid multi-core architectures, where the gap between processor and memory performance continues to grow. The speedup can be achieved by ordering unknowns so that ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015