Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2407.00656

Multiple-GPU accelerated high-order gas-kinetic scheme on three-dimensional unstructured meshes

Yuhang Wang·Waixiang Cao·Liang Pan
View original source ↗Read source document ↗Export metadata ↓

Abstract

Recently, successes have been achieved for the high-order gas-kinetic schemes (HGKS) on unstructured meshes for compressible flows. In this paper, to accelerate the computation, HGKS is implemented with the graphical processing unit (GPU) using the compute unified device architecture (CUDA). HGKS on unstructured meshes is a fully explicit scheme, and the acceleration framework can be developed based on the cell-level parallelism. For single-GPU computation, the connectivity of geometric information is generated for the requirement of data localization and independence. Based on such data structure, the kernels and corresponding girds of CUDA are set. With the one-to-one mapping between the indices of cells and CUDA threads, the single-GPU computation using CUDA can be implemented for HGKS. For multiple-GPU computation, the domain decomposition and data exchange need to be taken into account. The domain is decomposed into subdomains by METIS, and the MPI processes are created for the control of each process and communication among GPUs. With reconstruction of connectivity and adding ghost cells, the main configuration of CUDA for single-GPU can be inherited by each GPU. The benchmark cases for compressible flows, including accuracy test and flow passing through a sphere, are presented to assess the numerical performance of HGKS with Nvidia RTX A5000 and Tesla V100 GPUs. For single-GPU computation, compared with the parallel central processing unit (CPU) code running on the Intel Xeon Gold 5120 CPU with open multi-processing (OpenMP) directives, 5x speedup is achieved by RTX A5000 and 9x speedup is achieved by Tesla V100. For multiple-GPU computation, HGKS code scales properly with the increasing number of GPU. Numerical results confirm the excellent performance of multiple-GPU accelerated HGKS on unstructured meshes.

Explore related subjects

math.NAcs.NA

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Yuhang Wang, Waixiang Cao, Liang Pan. 2024-06-30. Multiple-GPU accelerated high-order gas-kinetic scheme on three-dimensional unstructured meshes. https://arxiv.org/abs/2407.00656

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

The Stability of Block Eliminations and Additive Modifications

The block elimination with additive modifications (BEAM) method was recently proposed as a alternative to LU with partial pivoting requiring less communication. Because of the novelty of BEAM, the existing theoretical analysis is lacking. To that end, we analyze both the numerical stability of the underlying block LU factorization and the effects of additive modifications. For the block LU factorization, we are able to improve the previous results of Demmel et al. from being cubic in the element growth to merely quadratic. Furthermore, we propose an alternative measure of element growth that is better aligned with block LU; this new measure of growth allows our analysis to apply to matrices that cannot be factored with pointwise LU. In the second part, we analyzed the modifications produced by BEAM and the effect they have on the condition number and growth factor. Finally, we show that BEAM will not apply any modifications in some cases that regular block LU can safely factor.

math.NA↗
arXiv2026

Efficient Rigorous Continuation via Chebyshev Series Expansion I

We study the global continuation of solution manifolds arising in dynamical systems. We present a rigorous continuation method based on a Chebyshev series expansion of the solution manifold. The branch is first approximated by a high-order Chebyshev interpolation polynomial, and an explicit error bound is then obtained by verifying the contraction of a quasi-Newton operator near this approximation. The contraction is formulated on a weighted $\ell^1$ space, giving a finer control than the typical $C^0$-error bound obtained from the uniform contraction theorem. In fact, the latter follows directly from our contraction operator. Furthermore, we discuss how our strategy applies naturally to pseudo-arclength continuation, where the continuation parameter fails to provide a valid local coordinate, and extends to multi-parameter continuation. Lastly, we detail two applications in which we compute a two-parameter family of steady-states for the Cahn--Hilliard equation, and a one-parameter family of steady-states undergoing saddle-node bifurcations for the Shigesada--Kawasaki--Teramoto system.

math.NA↗
arXiv2026

Efficient iterative techniques for solving tensor problems with the T-product

This paper develops two efficient iterative methods for solving tensor equations under the T-product framework. For T-symmetric positive definite tensor equations of the form $\mathcal{C} \star \mathcal{X} = \mathcal{D}$, we propose a conjugate-gradient-type algorithm that generates orthogonal residual and $\mathcal{C}$-orthogonal direction sequences, ensuring convergence within a finite number of steps. For general consistent tensor equations, we extend the method using a normal-equation transformation, and further adapt it to handle inconsistent systems by solving a least-squares minimization problem. Key advantages include direct tensor-based computations without explicit matrix expansion, rigorous finite-step convergence proofs, and the ability to obtain minimal Frobenius norm solutions. Numerical experiments on synthetic data, benchmark images, and video sequences demonstrate that the proposed algorithms achieve high precision with low computational time, confirming their practicality for large-scale multidimensional problems.

math.NA↗