arXiv subjects

Simone Bnà

Publications and source records attributed to Simone Bnà.

5 recordsLinked to original sources

arXiv2026

Accelerating Sparse Linear Solvers in OpenFOAM using RISC-V Vector Extensions

Computational Fluid Dynamics (CFD) relies heavily on the efficiency of linear solvers based on sparse linear algebra kernels. Widely used frameworks like OpenFOAM exploit parallelism primarily at the domain decomposition level via MPI. Support for vector/SIMD architectures is limited to compiler auto-vectorization. Furthermore, support for such architectures is limited by OpenFOAM's internal matrix data format, which is intrinsically ill-suited for the contiguous memory accesses required for efficient execution on vector processors. In this work, we focused on two very different RISC-V architectures: the prototype long-vector EPAC accelerator and the commercial short-vector CPU Sophon SG2044. On these platforms, we optimized the Sparse Matrix-Vector multiplication (SpMV) using RISC-V vector intrinsics and integrated it into a custom smoother, performing a runtime conversion of internal data into a vector-friendly format. Experimental results on the EPAC test chip show a 6x speedup for the smoother; benchmarks on Monte Cimone (MCv2) cluster with the Sophon SG2044 processor achieve a 1.5x smoother speedup, proving that legacy CFD codes can be effectively accelerated on both research and commercial emerging hardware.

cs.DC↗

arXiv2025

SPUMA: a minimally invasive approach to the GPU porting of OPENFOAM

High Performance Computing (HPC) on hybrid clusters represents a significant opportunity for Computational Fluid Dynamics (CFD), especially when modern accelerators are utilized effectively. However, despite the widespread adoption of GPUs, programmability remains a challenge, particularly in open-source contexts. In this paper, we present SPUMA, a full GPU porting of OPENFOAM targeting NVIDIA and AMD GPUs. The implementation strategy is based on a portable programming model and the adoption of a memory pool manager that leverages the unified memory feature of modern GPUs. This approach is discussed alongside several numerical tests conducted on two pre-exascale clusters in Europe, LUMI and Leonardo, which host AMD MI250X and NVIDIA A100 GPUs, respectively. In the performance analysis section, we present results related to memory usage profiling and kernel wall-time, the impact of the memory pool, and energy consumption obtained by simulating the well-known DrivAer industrial test case. GPU utilization strongly affects strong scalability results, reaching 65% efficiency on both LUMI and Leonardo when approaching a load of 8 million cells per GPU. Weak scalability results, obtained on 20 GPUs with the OpenFOAM native multigrid solver, range from 75% on Leonardo to 85% on LUMI. Notably, efficiency is no lower than 90% when switching to the NVIDIA AmgX linear algebra solver. Our tests also reveal that one A100 GPU on Leonardo is equivalent 200-300 Intel Sapphire Rapids cores, provided the GPUs are sufficiently oversubscribed (more than 10 million of cells per GPU). Finally, energy consumption is reduced by up to 82% compared to analogous simulations executed on CPUs.

cs.DC↗

arXiv2025

Coupled CFD-DEM model for dry powder inhalers simulation: validation and sensitivity analysis for the main model parameters

The use of computational techniques in the design of dry powder inhalers (DPI), as well as in unravelling the complex mechanisms of drug aerosolization, has increased significantly in recent years. Computational fluid dynamics (CFD) is used to study the air flow, inside the DPI, during the patient inspiratory act while discrete element methods (DEM) are used to simulate the dispersion and aerosolization of the drug product powder particles. In this work we discuss the possibility to validate a coupled CFD-DEM model for the NextHaler DPI device against previously published experimental data. The approximations and assumptions made are deeply discussed. The comparison between computational and experimental results is detailed both for fluid and powder flows. Finally, the potential and possible applications of a calibrated DPI model are discussed as well as the missing elements necessary to achieve a fully quantitatively predictive computational model.

physics.app-ph↗

arXiv2025

Investigation of particle dynamics and classification mechanism in a spiral jet mill through computational fluid dynamics and discrete element methods

Predicting the outcome of jet-milling based on the knowledge of process parameters and starting material properties is a task still far from being accomplished. Given the technical difficulties in measuring thermodynamics, flow properties and particle statistics directly in the mills, modelling and simulations constitute alternative tools to gain insight in the process physics and many papers have been recently published on the subject. An ideal predictive simulation tool should combine the correct description of non-isothermal, compressible, high Mach number fluid flow, the correct particle-fluid and particle-particle interactions and the correct fracture mechanics of particle upon collisions but it is not currently available. In this paper we present our coupled CFD-DEM simulation results; while comparing them with the recent modelling and experimental works we will review the current understating of the jet-mill physics and particle classification. Subsequently we analyze the missing elements and the bottlenecks currently limiting the simulation technique as well as the possible ways to circumvent them towards a quantitative, predictive simulation of jet-milling.

physics.comp-ph↗

arXiv2024

Porting HPC Applications to AMD Instinct$^\text{TM}$ MI300A Using Unified Memory and OpenMP

AMD Instinct$^\text{TM}$ MI300A is the world's first data center accelerated processing unit (APU) with memory shared between the AMD "Zen 4" EPYC$^\text{TM}$ cores and third generation CDNA$^\text{TM}$ compute units. A single memory space offers several advantages: i) it eliminates the need for data replication and costly data transfers, ii) it substantially simplifies application development and allows an incremental acceleration of applications, iii) is easy to maintain, and iv) its potential can be well realized via the abstractions in the OpenMP 5.2 standard, where the host and the device data environments can be unified in a more performant way. In this article, we provide a blueprint of the APU programming model leveraging unified memory and highlight key distinctions compared to the conventional approach with discrete GPUs. OpenFOAM, an open-source C++ library for computational fluid dynamics, is presented as a case study to emphasize the flexibility and ease of offloading a full-scale production-ready application on MI300 APUs using directive-based OpenMP programming.

cs.DC↗