Graphics Processing Units Research Articles

Tensor power flow formulations for multidimensional analyses in distribution systems

In this paper, we present two multidimensional power flow formulations based on a fixed-point iteration (FPI) algorithm to efficiently solve hundreds of thousands of Power flows (PFs) in distribution systems. The presented algorithms are the base for a new TensorPowerFlow (TPF) tool and shine for their simplicity, benefiting from multicore Central processing unit (CPU) and Graphics processing unit (GPU) parallelization. We also focus on the mathematical convergence properties of the algorithm, showing that its unique solution is at the practical operational point. The proof is validated using numerical simulations showing the robustness of the FPI algorithm compared to the classical Newton–Raphson (NR) approach. In the case study, a benchmark with different PF solution methods is performed, showing that for applications requiring a yearly simulation at 1-minute resolution, the computation time is decreased by a factor of 164, compared to the NR in its sparse formulation. Finally, a set of applications is described, highlighting the potential of the proposed formulations over a wide range of analyses in distribution systems.

Immersed boundary method for dynamic simulation of polarizable colloids of arbitrary shape in explicit ion electrolytes.

We develop a computational method for modeling electrostatic interactions of arbitrarily shaped, polarizable objects on colloidal length scales, including colloids/nanoparticles, polymers, and surfactants, dispersed in explicit ion electrolytes and nonionic solvents. Our method computes the nonuniform polarization charge distribution induced in a colloidal particle by both externally applied electric fields and local electric fields arising from other charged objects in the dispersion. This leads to expressions for electrostatic energies, forces, and torques that enable efficient molecular dynamics and Brownian dynamics simulations of colloidal dispersions in electrolytes, which can be harnessed to accurately predict structural and transport properties. We describe an implementation in which colloidal particles are modeled as rigid composites of small spherical beads that tessellate the surface of the particle. The electrostatics calculations are accelerated using a spectrally accurate particle-mesh-Ewald technique implemented on a graphics processing unit and regularized such that the electrostatic calculations are well-defined even for overlapping bodies. We illustrate the effectiveness of this approach with a comprehensive set of calculations: the induced dipole moments and forces for individual, paired, and lattice configurations of spherical colloids in an electric field; the induced dipole moment and torque for anisotropic particles subjected to an electric field; the equilibrium ion distribution in the double layer surrounding charged colloids; the dynamics of charged colloids; and the behavior of ions in the double layer of a polarizable colloid under the influence of an electric field.

Improved modularity and new features in ipie: Toward even larger AFQMC calculations on CPUs and GPUs at zero and finite temperatures.

ipie is a Python-based auxiliary-field quantum Monte Carlo (AFQMC) package that has undergone substantial improvements since its initial release [Malone et al., J. Chem. Theory Comput. 19(1), 109-121 (2023)]. This paper outlines the improved modularity and new capabilities implemented in ipie. We highlight the ease of incorporating different trial and walker types and the seamless integration of ipie with external libraries. We enable distributed Hamiltonian simulations of large systems that otherwise would not fit on a single central processing unit node or graphics processing unit (GPU) card. This development enabled us to compute the interaction energy of a benzene dimer with 84 electrons and 1512 orbitals with multi-GPUs. Using CUDA and cupy for NVIDIA GPUs, ipie supports GPU-accelerated multi-slater determinant trial wavefunctions [Huang et al. arXiv:2406.08314 (2024)] to enable efficient and highly accurate simulations of large-scale systems. This allows for near-exact ground state energies of multi-reference clusters, [Cu2O2]2+ and [Fe2S2(SCH3)4]2-. We also describe implementations of free projection AFQMC, finite temperature AFQMC, AFQMC for electron-phonon systems, and automatic differentiation in AFQMC for calculating physical properties. These advancements position ipie as a leading platform for AFQMC research in quantum chemistry, facilitating more complex and ambitious computational method development and their applications.

Fault3DNnet: A lightweight 3D seismic fault detection network with bidirectional decoding

Fault detection from seismic data is a critical component of hydrocarbon exploration and production. In recent years, deep-learning techniques have made tremendous achievements and breakthroughs in seismic fault detection. Especially U-nets, represented by FaultSeg3D and its variants, are a dominant method for 3D deep-learning-based seismic fault interpretation. The incorporation of state-of-the-art network modules or training strategies with U-nets continuously improves the performance of fault detection for field seismic data, usually at the cost of increased computational intensity and hardware consumption. We design a lightweight bidirectional decoding network, Fault3DNnet, for fault detection from 3D seismic data volume. To make it simple and concise, our Fault3DNnet is equipped with basic conventional convolution and regular operation modules. The two decoding branches in Fault3DNnet could provide complementary information and improve fault detection performance. The superiority of Fault3DNnet is verified using publicly available synthetic and four field seismic data sets. The experimental results indicate that the method predicts fault structures with improved completeness and continuity compared with FaultSeg3D. Meanwhile, the parameter and computational quantity of the network are only 1/7 and 1/5 of FaultSeg3D, respectively. With the same graphics processing unit memory consumption, Fault3DNnet can handle the size of approximately six times the volume of 3D seismic data during inferencing as that of FaultSeg3D. It helps to improve prediction stability and decrease discontinuous artifacts at the stitching boundaries when predicting the large field data in chunks. In general, our Fault3DNnet delivers superior fault detection results while greatly reducing computational intensity and memory usage.

High-throughput software LDPC decoder on GPU

The new or future communication systems require much higher flexibility and scalability to support diverse scenarios. In this paper, a high-throughput low-density parity-check (LDPC) decoder on graphics processing unit is presented to meet the flexible and scalable requirements. A memory-reduced forward/backward approach for the check node update is proposed. Moreover, elaborate on-chip memory allocations are conducted to improve memory bandwidth. The proposed (26112, 8448) 5G LDPC decoder on RTX4090 achieves 27.6 Gbps decoding throughput with five layered iterations, while the latency is less than 1 ms. Compared with related works, the throughput speedups obtained by the presented LDPC decoder are from 1.18×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$1.18\ imes$$\\end{document} to 12.4×\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$12.4\ imes$$\\end{document}.

Reducing the replication time for structural estimations: A successful replication of “An Anatomy of International Trade” using GPU computing

AbstractEaton, Kortum, and Kramarz (2011) (EKK) discovered empirical patterns from French manufacturing firms that a baseline firm heterogeneity model could not explain. The authors proposed and estimated a model that closely matches the patterns observed in French data. This paper successfully replicates their findings using author‐provided data, re‐implementing their algorithms in Python and leveraging graphics processing unit computing to significantly boost computational speed. Applying the model to Chinese manufacturing data, despite differences in context, showed that the model explains most observed patterns well.

Big data research is everyone's research-Making epilepsy data science accessible to the global community: Report of the ILAE big data commission.

Epilepsy care generates multiple sources of high-dimensional data, including clinical, imaging, electroencephalographic, genomic, and neuropsychological information, that are collected routinely to establish the diagnosis and guide management. Thanks to high-performance computing, sophisticated graphics processing units, and advanced analytics, we are now on the cusp of being able to use these data to significantly improve individualized care for people with epilepsy. Despite this, many clinicians, health care providers, and people with epilepsy are apprehensive about implementing Big Data and accompanying technologies such as artificial intelligence (AI). Practical, ethical, privacy, and climate issues represent real and enduring concerns that have yet to be completely resolved. Similarly, Big Data and AI-related biases have the potential to exacerbate local and global disparities. These are highly germane concerns to the field of epilepsy, given its high burden in developing nations and areas of socioeconomic deprivation. This educational paper from the International League Against Epilepsy's (ILAE) Big Data Commission aims to help clinicians caring for people with epilepsy become familiar with how Big Data is collected and processed, how they are applied to studies using AI, and outline the immense potential positive impact Big Data can have on diagnosis and management.

An explicit time integration method for Boussinesq approximation

AbstractThis paper presents a novel approach to the discretization of the Boussinesq equation using finite elements, coupled with an explicit pressure correction scheme. The entire numerical scheme is formulated exclusively in terms of matrix‐vector products, enabling a highly efficient numerical solution. The key advantage of this approach lies in its suitability for fast computations and seamless parallelization on graphics processing units (GPUs). We present the numerical scheme and discuss different computational examples.

An artifactual fibre overlap removal algorithm for micro-computed tomography image post-processing and 3D microstructure generation with graphics processing unit acceleration

A novel algorithm based on radial basis functions is proposed for the removal of artifactual fibre overlap within fibre structures extracted from micro-computed tomography (micro-CT) images of fibre reinforced polymer matrix composites. The proposed algorithm is highly efficient and excels in preserving the original fibre structures extracted from the micro-CT images. Besides, graphics processing unit (GPU) acceleration is applied to further enhance the efficiency of the fibre overlap removal process. Furthermore, the proposed algorithm is also modified for the generation of periodic 3D microstructures. For practical application, the proposed algorithm is implemented for both the artifactual fibre overlap removal within micro-CT images from an injection moulded part and the microstructure generation for 3D printed samples. The unidirectional elastic modulus of the resultant microstructures is computed via numerical simulations and shows a close match to the experimental measurements with relative errors less than 2%. Overall, the proposed algorithm significantly facilitates the reconstruction of micro-CT image-based numerical models and can also be easily repurposed to generate complex microstructures, which is of great value for the development of data-driven models for characterization and design of composite materials that demands large amounts of data on material microstructures.

All-to-all reconfigurability with sparse and higher-order Ising machines

Domain-specific hardware to solve computationally hard optimization problems has generated tremendous excitement. Here, we evaluate probabilistic bit (p-bit) based Ising Machines (IM) on the 3-Regular 3-Exclusive OR Satisfiability (3R3X), as a representative hard optimization problem. We first introduce a multiplexed architecture that emulates all-to-all network functionality while maintaining highly parallelized chromatic Gibbs sampling. We implement this architecture in a single Field-Programmable Gate Array (FPGA) and show that running the adaptive parallel tempering algorithm demonstrates competitive algorithmic and prefactor advantages over alternative IMs by D-Wave, Toshiba, and Fujitsu. We also implement higher-order interactions that lead to better prefactors without changing algorithmic scaling for the XORSAT problem. Even though FPGA implementations of p-bits are still not quite as fast as the best possible greedy algorithms accelerated on Graphics Processing Units (GPU), scaled magnetic versions of p-bit IMs could lead to orders of magnitude improvements over the state of the art for generic optimization.

Evaluating Medical Entity Recognition in Health Care: Entity Model Quantitative Study.

Named entity recognition (NER) models are essential for extracting structured information from unstructured medical texts by identifying entities such as diseases, treatments, and conditions, enhancing clinical decision-making and research. Innovations in machine learning, particularly those involving Bidirectional Encoder Representations From Transformers (BERT)-based deep learning and large language models, have significantly advanced NER capabilities. However, their performance varies across medical datasets due to the complexity and diversity of medical terminology. Previous studies have often focused on overall performance, neglecting specific challenges in medical contexts and the impact of macrofactors like lexical composition on prediction accuracy. These gaps hinder the development of optimized NER models for medical applications. This study aims to meticulously evaluate the performance of various NER models in the context of medical text analysis, focusing on how complex medical terminology affects entity recognition accuracy. Additionally, we explored the influence of macrofactors on model performance, seeking to provide insights for refining NER models and enhancing their reliability for medical applications. This study comprehensively evaluated 7 NER models-hidden Markov models, conditional random fields, BERT for Biomedical Text Mining, Big Transformer Models for Efficient Long-Sequence Attention, Decoding-enhanced BERT with Disentangled Attention, Robustly Optimized BERT Pretraining Approach, and Gemma-across 3 medical datasets: Revised Joint Workshop on Natural Language Processing in Biomedicine and its Applications (JNLPBA), BioCreative V CDR, and Anatomical Entity Mention (AnatEM). The evaluation focused on prediction accuracy, resource use (eg, central processing unit and graphics processing unit use), and the impact of fine-tuning hyperparameters. The macrofactors affecting model performance were also screened using the multilevel factor elimination algorithm. The fine-tuned BERT for Biomedical Text Mining, with balanced resource use, generally achieved the highest prediction accuracy across the Revised JNLPBA and AnatEM datasets, with microaverage (AVG_MICRO) scores of 0.932 and 0.8494, respectively, highlighting its superior proficiency in identifying medical entities. Gemma, fine-tuned using the low-rank adaptation technique, achieved the highest accuracy on the BioCreative V CDR dataset with an AVG_MICRO score of 0.9962 but exhibited variability across the other datasets (AVG_MICRO scores of 0.9088 on the Revised JNLPBA and 0.8029 on AnatEM), indicating a need for further optimization. In addition, our analysis revealed that 2 macrofactors, entity phrase length and the number of entity words in each entity phrase, significantly influenced model performance. This study highlights the essential role of NER models in medical informatics, emphasizing the imperative for model optimization via precise data targeting and fine-tuning. The insights from this study will notably improve clinical decision-making and facilitate the creation of more sophisticated and effective medical NER models.

Large scale multi-GPU based parallel traffic simulation for accelerated traffic assignment and propagation

Traffic simulation is a critical tool for congestion analysis, travel time estimation, and route optimization in urban planning, benefiting navigation apps, transportation network companies, and state agencies. Traditionally, traffic micro-simulation frameworks are based on road segments and can only support a limited number of main roads. Efficient traffic simulation on a regional scale remains a significant challenge due to the complexity of urban mobility and the large scale of spatiotemporal data. This paper introduces a Large Scale Multi-GPU Parallel Computing based Regional Scale Traffic Simulation Framework (LPSim), which leverages graphical processing unit (GPU) parallel computing to address these challenges. LPSim utilizes a multi-GPU architecture to simulate extensive and dynamic traffic networks with high fidelity and reduced computation time. Using the parallel processing capabilities of GPUs, LPSim can perform tens of millions of individual vehicle dynamics simulations simultaneously, significantly outperforming traditional CPU-based approaches. The framework is designed to be scalable and can easily accommodate the increasing complexity of traffic simulations. We present the theory behind GPU-based traffic simulation, the architecture of single- and multi-GPU based simulations, and the graph partition strategies that enhance computation resource load balance. Our experimental results demonstrate the effectiveness of LPSim in simulating large-scale traffic scenarios. LPSim is capable of completing simulations of 2.82 million trips in just 6.28 min on a single GPU machine equipped with 5120 CUDA cores (Tesla V100-SXM2). Furthermore, utilizing a Google Cloud instance with two NVIDIA V100 GPUs, which collectively offer 10240 CUDA cores, LPSim successfully simulates 9.01 million trips within 21.16 min. We further tested our simulator with the same demand on dual NVIDIA A100-PCIE-40GB GPUs, which finished the simulation in 0.0398 h, approximately 113 times faster than the same simulation scenario running on an Intel(R) Xeon(R) Gold 6326 CPU @ 2.90 GHz, which takes 4.49 h to complete. This performance not only demonstrates its speed and scalability advantages over traditional simulation techniques but also highlights LPSim’s unique position as the first traffic simulation framework that is scalable for both single- and multiple-GPU configurations. Consequently, LPSim provides an invaluable tool for individuals and extensive research teams alike, enabling the acquisition of large-scale traffic simulation results in a time-efficient manner. LPSim code is available at: https://github.com/Xuan-1998/LPSim

HybridSA: GPU Acceleration of Multi-pattern Regex Matching using Bit Parallelism

Multi-pattern matching is widely used in modern software for applications requiring high throughput such as protein search, network traffic inspection, virus or spam detection. Graphics Processor Units (GPUs) excel at executing massively parallel workloads. Regular expression (regex) matching is typically performed by simulating the execution of deterministic finite automata (DFAs) or nondeterministic finite automata (NFAs). The natural implementations of these automata simulation algorithms on GPUs are highly inefficient because they give rise to irregular memory access patterns. This paper presents HybridSA, a heterogeneous CPU-GPU parallel engine for multi-pattern matching. HybridSA uses bit parallelism to efficiently simulate NFAs on GPUs, thus reducing the number of memory accesses and increasing the throughput. Our bit-parallel algorithms extend the classical shift-and algorithm for string matching to a large class of regular expressions and reduce automata simulation to a small number of bitwise operations. We have developed a compiler to translate regular expressions into bit masks, perform optimizations, and choose the best algorithms to run on the GPU. The majority of the regular expressions are accelerated on the GPU, while the patterns that exhibit random memory accesses are executed on the CPU in parallel. We evaluate HybridSA against state-of-the-art CPU and GPU engines, as well as a hybrid combination of the two. HybridSA achieves between 4 and 60 times higher throughput than the state-of-the-art CPU engine and between 4 and 233 times better than the state-of-the-art GPU engine across a collection of real-world benchmarks.

Accelerating Neural Network Inference in Handwritten Digit Recognition — Comparative Study

In deep learning, one of the popular subclasses of multilayer perceptron is the Convolutional Neural Networks (CNNs), optimized for image and semantic segmentation tasks. With a large number of floating-point operations and relatively less data transfer in the training phase, CNNs are well suited to be handled by parallel architectures. The high accuracy of CNNs comes at the cost of consequential compute and memory demands. In this article, we present a comparative analysis of a pre-trained CNN framework for recognizing a handwritten digit on different processing platforms. The performance of the neural network as the function of the parallel processing platforms offers parametric insights and factors that influence the inference on state-of-the-art Graphics Processing Units (GPUs) and systolic arrays such as Eyeriss and Tensor Processing Unit (TPU). Through inference time analysis, we observed that the systolic arrays can outperform upto 58.7 times better than a Turing architecture powered GPU. We show that while the efficiency with which the available resources are utilized is higher in the existing GPUs (upto 32%) than the TPUs, the efficiency of application-specific systolic arrays can be on par with that of GPUs. We present the results obtained from three different customized systolic array-based platforms which can be adopted by the designers to decide the hardware optimization goal.

A GPU-Based Lattice Boltzmann Method for Predicting Near- and Far-Field Jet Noise

Jet flow and noise are simulated using the lattice Boltzmann method (LBM). The D3Q19 lattice model calculates the flow field with a high-order distribution function to overcome LBM limitations in solving high Mach number flows. Multiple relaxation times (MRT) and large eddy simulation (LES) are used in this study. Compressible LBM was applied to the simulation jet flow to demonstrate the ability of the fifth-order equilibrium distribution function to simulate compressible flows efficiently. The results from the LBM simulation were then employed in Kirchhoff's surface integral approach to predict far-field jet noise. In this research, we used the compute unified device architecture (CUDA) to implement the LBM in a graphics processing unit (GPU), thus creating a hybrid code, LBM-MRT-LES. The obtained results agreed with the experimental and numerical results in the literature.

Very-Large-Scale GPU-Accelerated Nuclear Gradient of Time-Dependent Density Functional Theory with Tamm-Dancoff Approximation and Range-Separated Hybrid Functionals.

Modern graphics processing units (GPUs) provide an unprecedented level of computing power. In this study, we present a high-performance, multi-GPU implementation of the analytical nuclear gradient for Kohn-Sham time-dependent density functional theory (TDDFT), employing the Tamm-Dancoff approximation (TDA) and Gaussian-type atomic orbitals as basis functions. We discuss GPU-efficient algorithms for the derivatives of electron repulsion integrals and exchange-correlation functionals within the range-separated scheme. As an illustrative example, we calculate the TDA-TDDFT gradient of the S1 state of a full-scale green fluorescent protein with explicit water solvent molecules, totaling 4353 atoms, at the ωB97X/def2-SVP level of theory. Our algorithm demonstrates favorable parallel efficiencies on a high-speed distributed system equipped with 256 Nvidia A100 GPUs, achieving >70% with up to 64 GPUs and 31% with 256 GPUs, effectively leveraging the capabilities of modern high-performance computing systems.

Efficient Random Field Generation With Rotational Anisotropy for Probabilistic SPH Analysis of Slope Failure

ABSTRACTDue to geological processes such as sedimentation, tectonic movement, and backfilling, natural soil often exhibits characteristics of rotated anisotropy. Recent studies have shown the significant impact of rotated anisotropy on slope stability. However, little research has explored how this rotated anisotropy affects the large deformations occurring after slope failure. Therefore, this study integrates rotated random field theory with smoothed particle hydrodynamics (SPH) to investigate its influence on post‐failure slope behavior. Focusing on a typical slope scenario, this research utilizes graphics processing unit (GPU)–accelerated covariance matrix decomposition (CMD) method to create rotated anisotropy random fields and applies the SPH framework for analysis. It examines the influence of rotated anisotropy angles and the cross‐correlation between cohesion and internal friction angle on landslides. The results indicate that the rotational anisotropy of the slope significantly influences post‐failure behavior. When the rotation angle is close to the slope surface, it tends to amplify both the magnitude and variability of slope failure. Furthermore, the study evaluates the efficiency of generating these random fields and emphasizes the substantial computational speed improvements achieved with GPU acceleration. These findings offer a robust approach for probabilistic analysis of slope large deformations considering rotated anisotropy. They provide a theoretical foundation for accurately assessing the risk of slope collapse, holding significant practical implications for geotechnical engineering.

Dynamic Load Balancing of Multi-GPU Parallelization for VULCANO VE-U7 Corium Spreading Analysis Using SOPHIA

The SOPHIA code is a graphical processing unit (GPU)–based smoothed particle hydrodynamics (SPH) framework for fundamental nuclear thermal-hydraulic and safety applications developed by Seoul National University. Focusing on particle-based SPH methods, the SOPHIA code provides capabilities for computational fluid dynamics–level analysis in addressing complex multiphase and multiphysics issues related to severe accidents and reactor safety. Since the SPH method requires high computational cost due to the nature of the particle-based method, the SOPHIA code was parallelized with multiple GPUs. However, using multiple GPUs may cause load imbalance as particles migrate between GPUs, which degrades performance. In this study, we developed a dynamic load-balancing algorithm to increase computational efficiency by reducing the load imbalance. Additionally, a novel cumulative time comparison approach is proposed as a criterion for adjusting boundaries between GPUs, ensuring low load imbalance. To evaluate the performance of the algorithm, three-dimensional dam-break simulations were conducted. The results demonstrated enhanced computational efficiency and equitable load distribution among GPUs compared to using fixed subdomains. Additionally, the developed algorithm was utilized to analyze the VULCANO VE-U7 experiment, and the results were well aligned with the experimental data. Therefore, the practical applicability of the proposed dynamic load-balancing algorithm in nuclear safety analyses has been confirmed.

GPU implementation of a fast active noise control algorithm with multiple reference signals

Traditional active noise control (ANC) systems are mostly implemented on digital signal processor (DSP). Due to the inadequate computational power of the DSP, especially in applications with multiple reference signals, the DSP is unable to update the optimal control filter for different types of noise. Hence, Graphics Processing Unit (GPU) has been incorporated to the ANC system to enhance its computational power. In this work, a DSP-CPU-GPU computing architecture is adopted to improve the computational performance of the ANC system and ensure the real-time signal processing. In addition, a fast ANC algorithm on the GPU is employed to update the control filters to handle different reference signals. Finally, the effectiveness of the method is verified by experiments in the condition of multiple reference signals.

Direct aeroacoutic simulation of a ducted axial fan with acoustic liners using a volume penalization lattice Boltzmann solver on heterogeneous massively parallel platforms

Aeroacoustic noise of an axial fan is directly solved to understand its mechanism, using our in-house Lattice Boltzmann Method (LBM) solver. To make the LBM solver industrially successful, we must reduce calculation time while maintaining its accuracy. We adopt Volume Penalization method for moving boundaries to achieve independent parallelism, and to utilize GPU (Graphical Processing Units) for parallel computation. In this study, we present a comparison between numerical and experimental results of an axial fan in a lined duct. The acoustic liner is designed to reduce 2nd BPF noise, and the noise reduction of 9.8dB is observed in the experiment. With our LBM code, the noise reduction was reproduced within an accuracy of 2.5dB, and profound insights are provided through the detailed investigation of the numerical results.