Graphics Processing Units Processing Research Articles

An Implementation of LASER Beam Welding Simulation on Graphics Processing Unit Using CUDA

The maximum number of parallel threads in traditional CFD solutions is limited by the Central Processing Unit (CPU) capacity, which is lower than the capabilities of a modern Graphics Processing Unit (GPU). In this context, the GPU allows for simultaneous processing of several parallel threads with double-precision floating-point formatting. The present study was focused on evaluating the advantages and drawbacks of implementing LASER Beam Welding (LBW) simulations using the CUDA platform. The performance of the developed code was compared to that of three top-rated commercial codes executed on the CPU. The unsteady three-dimensional heat conduction Partial Differential Equation (PDE) was discretized in space and time using the Finite Volume Method (FVM). The Volumetric Thermal Capacitor (VTC) approach was employed to model the melting-solidification. The GPU solutions were computed using a CUDA-C language in-house code, running on a Gigabyte Nvidia GeForce RTX™ 3090 video card and an MSI 4090 video card (both made in Hsinchu, Taiwan), each with 24 GB of memory. The commercial solutions were executed on an Intel® Core™ i9-12900KF CPU (made in Hillsboro, Oregon, United States of America) with a 3.6 GHz base clock and 16 cores. The results demonstrated that GPU and CPU processing achieve similar precision, but the GPU solution exhibited significantly faster speeds and greater power efficiency, resulting in speed-ups ranging from 75.6 to 1351.2 times compared to the CPU solutions. The in-house code also demonstrated optimized memory usage, with an average of 3.86 times less RAM utilization. Therefore, adopting parallelized algorithms run on GPU can lead to reduced CFD computational costs compared to traditional codes while maintaining high accuracy.

Graphics-processing-unit-accelerated ice flow solver for unstructured meshes using the Shallow-Shelf Approximation (FastIceFlo v1.0.1)

Abstract. Ice-sheet flow models capable of accurately projecting their future mass balance constitute tools to improve flood risk assessment and assist sea-level rise mitigation associated with enhanced ice discharge. Some processes that need to be captured, such as grounding-line migration, require high spatial resolution (under the kilometer scale). Conventional ice flow models mainly execute on central processing units (CPUs), which feature limited parallel processing capabilities and peak memory bandwidth. This may hinder model scalability and result in long run times, requiring significant computational resources. As an alternative, graphics processing units (GPUs) are ideally suited for high spatial resolution, as the calculations can be performed concurrently by thousands of threads, processing most of the computational domain simultaneously. In this study, we combine a GPU-based approach with the pseudo-transient (PT) method, an accelerated iterative and matrix-free solution strategy, and investigate its performance for finite elements and unstructured meshes with application to two-dimensional (2-D) models of real glaciers at a regional scale. For both the Jakobshavn and Pine Island glacier models, the number of nonlinear PT iterations required to converge a given number of vertices (N) scales in the order of 𝒪(N1.2) or better. We further compare the performance of the PT CUDA C implementation with a standard finite-element CPU-based implementation using the price-to-performance metric. The price of a single Tesla V100 GPU is 1.5 times that of two Intel Xeon Gold 6140 CPUs. We expect a minimum speedup of at least 1.5 times to justify the Tesla V100 GPU price to performance. Our developments result in a GPU-based implementation that achieves this goal with a speedup beyond 1.5 times. This study represents a first step toward leveraging GPU processing power, enabling more accurate polar ice discharge predictions. The insights gained will benefit efforts to diminish spatial resolution constraints at higher computing performance. The higher computing performance will allow for ensembles of ice-sheet flow simulations to be run at the continental scale and higher resolution, a previously challenging task. The advances will further enable the quantification of model sensitivity to changes in upcoming climate forcings. These findings will significantly benefit process-oriented sea-level-projection studies over the coming decades.

GPU acceleration of conjugate gradient method obtaining Green's function for transport-property calculation

We present a graphics processing unit (GPU) acceleration procedure for the transport-property calculation of the RSPACE code, which uses density functional theory and the real-space finite-difference method. Porting the RSPACE code to run on a GPU using Open Accelerator and libraries results in the improvement of the computational speed and efficiency compared with a central processing unit. Furthermore, we achieved a speed up of ∼ 4x compared with a single GPU and effective processing in a multi-GPU environment by parallelizing in a way that minimizes communication and synchronization. Our GPU-acceleration procedure is demonstrated by the calculation of electron-transport properties of (4,4) carbon nanotubes (CNTs) with a 5-8-5 defect for which the calculations are suitable for parallel processing on GPUs. It is found that the information loss is strongly affected by the defect geometry when electron waves propagating in CNTs are utilized as the information transfer signal.

A Heterogeneous Parallel Algorithm for Euler-Lagrange Simulations of Liquid in Supersonic Flow

In spite of its prevalent usage for simulating the full-field process of the two-phase flow, the Euler–Lagrange method suffers from a heavy computing burden. Graphics processing units (GPUs), with their massively parallel architecture and high floating-point performance, provide new possibilities for high-efficiency simulation of liquid-jet-related systems. In this paper, a central processing unit/graphics processing unit (CPU/GPU) parallel algorithm based on the Euler–Lagrange scheme is established, in which both the gas and liquid phase are executed on the GPUs. To realize parallel tracking of the Lagrange droplets, a droplet dynamic management method is proposed, and a droplet-locating method is developed to address the cell. Liquid-jet-related simulations are performed on one core of the CPU with a GPU. The numerical results are consistent with the experiment. Compared with a setup using 32 cores of CPUs, considerable speedup is obtained, which is as high as 32.7 though it decreases to 20.2 with increasing droplets.

The implementation of the three-dimensional unified gas-kinetic wave-particle method on multiple graphics processing units

To further improve the efficiency of the unified gas-kinetic wave-particle (UGKWP) method in hypersonic rarefied non-equilibrium flows, particularly the particle simulation process, we presented the first application of the three-dimensional UGKWP method to multiple graphics processing unit (GPU) devices in this study. The wave and particle evolution components of the method are addressed using cell and particle paralleling strategies, respectively, enabling the primary loop of the GPU-based UPKWP (GPU-UGKWP) to be executed entirely by the compute unified device architecture threads on GPU devices. Concurrently, communication issues between central processing unit (CPU) nodes are resolved by employing the message passing interface model. Additionally, we introduce a tailored memory management scheme for the GPU-UGKWP method, facilitating efficient access to the particle array. Performance comparisons reveal that, relative to a single Intel Xeon Gold 6148 CPU core, the Nvidia Tesla P100 achieves a total speedup of 34 using one GPU device, and 226 with eight GPU devices, and a single Nvidia Titan V GPU device attains a speedup of 62. The speedup outcomes on multiple CPU cores and GPU devices demonstrate that the GPU-based algorithm is better suited for computationally demanding tasks, particularly in particle-dominated simulations. As evidenced by the reduced calculation time for a hypersonic technology vehicle simulation performed on the P100 cluster, GPU devices significantly outperform their CPU counterparts.

A Hybrid GPU and CPU Parallel Computing Method to Accelerate Millimeter-Wave Imaging

The range migration algorithm (RMA) based on Fourier transformation is widely applied in millimeter-wave (MMW) close-range imaging because of its few operations and small approximation. However, its interpolation stage is not effective due to the involved intensive logic controls, which limits the speed performance in a graphics processing unit (GPU) platform. Therefore, in this paper, we present an acceleration optimization method based on the hybrid GPU and central processing unit (CPU) parallel computation for implementing the RMA. The proposed method exploits the strong logic-control capability of the CPU to assist the GPU in processing the logic controls of the interpolation stage. The common positions of wavenumber-domain components to be interpolated are calculated by the CPU and stored in the constant memory for broadcast at any time. This avoids the repetitive computation consumed in a GPU-only scheme. Then the GPU is responsible for the remaining matrix-related steps and outputs the needed wavenumber-domain values. The imaging experiments verify the acceleration efficiency of the proposed method and demonstrate that the speedup ratio of our proposed method is more than 15 times of that by the CPU-only method, and more than 2 times of that by the GPU-only method.

Implementation of Beeman's algorithm to calculate execution time on GPU using CUDA

Graphics processing units (GPUs) are microprocessors designed to the operation of display and manipulation of graphics data. . Currently, these graphics processor are found on all graphics hardware and have become very important instruments for parallel computing. GPUs are practical tools for the development of several fields like decoding and encoding, solving differential equations. Their advantages are increase in performance, faster data processing and reduced power consumption. It is simple to program a GPU with CUDA C to run parallel calculations. But it is necessary to have an understanding of the architectural aspects of the GPU and CUDA C. This paper, we will describe and implement Beeman's algorithm on GPU and CPU using CUDA C to solve the differential equation of charged particles in an electromagnetic field. Our goal is to evaluate the performances of the implementation on GPU and CPU processors and to deduce the efficiency of the use of GPUs.

Sentence-Level Sentiment Classification A Comparative Study Between Deep Learning Models

Sentiment classification provides a means of analysing the subjective information in the text and subsequently extracting the opinion. Sentiment analysis is the method by which people extract information from their opinions, judgments and emotions about entities. In this paper we propose a comparative study between the most deep learning models used in the field of sentiment analysis; L-NFS (Linguistique Neuro Fuzzy System), GRU (Gated Recurrent Unit), BiGRU (Bidirectional Gated Recurrent Unit), LSTM (Long Short-Term Memory), BiLSTM (Bidirectional Long Short-Term Memory) and BERT(Bidirectional Encoder Representation from Transformers), we used for this study a large Corpus contain 1.6 Million tweets, as devices we train our models with GPU (graphics processing unit) processor. As result we obtain the best Accuracy and F1-Score respectively 87.36% and 0.87 for the BERT Model.

Analysis of Ionicity-Magnetism Competition in 2D-MX3 Halides towards a Low-Dimensional Materials Study Based on GPU-Enabled Computational Systems.

The acceleration of parallel high-throughput first-principle calculations in the context of 3D (three dimensional) periodic boundary conditions for low-dimensional systems, and particularly 2D materials, is an important issue for new material design. Where the scalability rapidly deflated due to the use of large void unit cells along with a significant number of atoms, which should mimic layered structures in the vacuum space. In this report, we explored the scalability and performance of the Quantum ESPRESSO package in the hybrid central processing unit - graphics processing unit (CPU-GPU) environment. The study carried out in the comparison to CPU-based systems for simulations of 2D magnets where significant improvement of computational speed was achieved based on the IBM ESSL SMP CUDA library. As an example of physics-related results, we have computed and discussed the ionicity-covalency and related ferro- (FM) and antiferro-magnetic (AFM) exchange competitions computed for some CrX compounds. Further, it has been demonstrated how this exchange interplay leads to high-order effects for the magnetism of the 1L-RuCl compound.

A graphics processing unit-based robust numerical model for solute transport driven by torrential flow condition

Solute transport simulations are important in water pollution events. This paper introduces a finite volume Godunov-type model for solving a 4×4 matrix form of the hyperbolic conservation laws consisting of 2D shallow water equations and transport equations. The model adopts the Harten-Lax-van Leer-contact (HLLC)-approximate Riemann solution to calculate the cell interface fluxes. It can deal well with the changes in the dry and wet interfaces in an actual complex terrain, and it has a strong shock-wave capturing ability. Using monotonic upstream-centred scheme for conservation laws (MUSCL) linear reconstruction with finite slope and the Runge-Kutta time integration method can achieve second-order accuracy. At the same time, the introduction of graphics processing unit (GPU)-accelerated computing technology greatly increases the computing speed. The model is validated against multiple benchmarks, and the results are in good agreement with analytical solutions and other published numerical predictions. The third test case uses the GPU and central processing unit (CPU) calculation models which take 3.865 s and 13.865 s, respectively, indicating that the GPU calculation model can increase the calculation speed by 3.6 times. In the fourth test case, comparing the numerical model calculated by GPU with the traditional numerical model calculated by CPU, the calculation efficiencies of the numerical model calculated by GPU under different resolution grids are 9.8–44.6 times higher than those by CPU. Therefore, it has better potential than previous models for large-scale simulation of solute transport in water pollution incidents. It can provide a reliable theoretical basis and strong data support in the rapid assessment and early warning of water pollution accidents.

Comparative study of the implementation of the Lagrange interpolation algorithm on GPU and CPU using CUDA to compute the density of a material at different temperatures

Graphics Processing Units (GPUs) are microprocessors attached to graphics cards, which are dedicated to the operation of displaying and manipulating graphics data. Currently, such graphics cards (GPUs) occupy all modern graphics cards. In a few years, these microprocessors have become potent tools for massively parallel computing. Such processors are practical instruments that serve in developing several fields like image processing, video and audio encoding and decoding, the resolution of a physical system with one or more unknowns. Their advantages: faster processing and consumption of less energy than the power of the central processing unit (CPU). In this paper, we will define and implement the Lagrange polynomial interpolation method on GPU and CPU to calculate the sodium density at different temperatures Ti using the NVIDIA CUDA C parallel programming model. It can increase computational performance by harnessing the power of the GPU. The objective of this study is to compare the performance of the implementation of the Lagrange interpolation method on CPU and GPU processors and to deduce the efficiency of the use of GPUs for parallel computing.

GPU simulation with Opticks: The future of optical simulations for LZ

The LZ collaboration aims to directly detect dark matter by using a liquid xenon Time Projection Chamber (TPC). In order to probe the dark matter signal, observed signals are compared with simulations that model the detector response. The most computationally expensive aspect of these simulations is the propagation of photons in the detector’s sensitive volume. For this reason, we propose to offload photon propagation modelling to the Graphics Processing Unit (GPU), by integrating Opticks into the LZ simulations workflow. Opticks is a system which maps Geant4 geometry and photon generation steps to NVIDIA’s OptiX GPU raytracing framework. This paradigm shift could simultaneously achieve a massive speed-up and an increase in accuracy for LZ simulations. By using the technique of containerization through Shifter, we will produce a portable system to harness the NERSC supercomputing facilities, including the forthcoming Perlmutter supercomputer, and enable the GPU processing to handle different detector configurations. Prior experience with using Opticks to simulate JUNO indicates the potential for speed-up factors over 1000× for LZ, and by extension other experiments requiring photon propagation simulations.

Graph Reachability on Parallel Many-Core Architectures

Many modern applications are modeled using graphs of some kind. Given a graph, reachability, that is, discovering whether there is a path between two given nodes, is a fundamental problem as well as one of the most important steps of many other algorithms. The rapid accumulation of very large graphs (up to tens of millions of vertices and edges) from a diversity of disciplines demand efficient and scalable solutions to the reachability problem. General-purpose computing has been successfully used on Graphics Processing Units (GPUs) to parallelize algorithms that present a high degree of regularity. In this paper, we extend the applicability of GPU processing to graph-based manipulation, by re-designing a simple but efficient state-of-the-art graph-labeling method, namely the GRAIL (Graph Reachability Indexing via RAndomized Interval) algorithm, to many-core CUDA-based GPUs. This algorithm firstly generates a label for each vertex of the graph, then it exploits these labels to answer reachability queries. Unfortunately, the original algorithm executes a sequence of depth-first visits which are intrinsically recursive and cannot be efficiently implemented on parallel systems. For that reason, we design an alternative approach in which a sequence of breadth-first visits substitute the original depth-first traversal to generate the labeling, and in which a high number of concurrent visits is exploited during query evaluation. The paper describes our strategy to re-design these steps, the difficulties we encountered to implement them, and the solutions adopted to overcome the main inefficiencies. To prove the validity of our approach, we compare (in terms of time and memory requirements) our GPU-based approach with the original sequential CPU-based tool. Finally, we report some hints on how to conduct further research in the area.

Numerical solution of three-dimensional time-dependent Schrödinger equation based on graphic processing unit acceleration

In the field of quantum mechanics, the theoretical study of the interaction between intense laser field and atoms and molecules depends very much on the numerical solution of the time-dependent Schrödinger equation. However, solving the three-dimensional time-dependent Schrödinger equation is not a simple task, and the analytical solution cannot be obtained, so it can only be solved numerically with the help of computer. In order to shorten the computing time and obtain the results quickly, it is necessary to use parallel methods to speed up computing. In this paper, under the background of strong field ionization, the three-dimensional time-dependent Schrödinger equation of hydrogen atom is solved in parallel, and the suprathreshold ionization of hydrogen atom under the action of linearly polarized infrared laser electric field is taken for example. Based on the spherical polar coordinate system, the time-dependent Schrödinger equation is discretized by the splitting operator-Fourier transform method, and the photoelectron continuous state wave function under the length gauge can be obtained. In Graphics processing unit (GPU) accelerated applications, the sequential portion of the workload runs on central processing unit (CPU) (which is optimized for single-threaded performance), while the compute-intensive part of the application runs in parallel on thousands of GPU cores. The GPU can make full use of the advantage of fine-grained parallelism based on multi-thread structure to realize parallel acceleration of the whole algorithm. Two accelerated computing modes of CPU parallel and GPU parallel are adopted, and their parallel acceleration performance is discussed. Compared with the results from the existing physical laws, the calculation error is also within an acceptable range, and the result is also consistent with the result from the existing physical laws of suprathreshold ionization, which also verifies the correctness of the program. In order to obtain a relatively accurate acceleration ratio, many different experiments are carried out. Computational experiments show that under the condition of ensuring accuracy, the GPU parallel computing speeds by up to about 60 times maximally based on the computational performance of CPU. It can be seen that the accelerated numerical solution of three-dimensional time-dependent Schrödinger equation based on GPU can significantly shorten the computational time. This work has important guiding significance for rapidly solving the three-dimensional time-dependent Schrödinger equation by using GPU.

A Performance Study of Moving Particle Semi-Implicit Method for Incompressible Fluid Flow on GPU

The aim of moving particle semi-implicit (MPS) is to simulate the incompressible flow of fluids in free surface. MPS, when implemented, consumes a lot of time and thus, needs a very powerful computing system. Instead of using parallel computing system, the performance level of the MPS model can be improved by using graphics processing units (GPUs). The aim is to have a computing system that is capable of performing at high levels thereby enhancing the speed of processing the numerical computations required in MPS. The primary aim of the study is to build a GPU-accelerated MPS model using CUDA aimed at reducing the time taken to perform the search for neighboring particles. In order to increase the GPU processing speed, specific consideration is given towards the optimization of a neighboring particle search process. The numerical model of MPS is performed using the governing equations, notably the Navier-Stokes equation. The simulation model indicates that using GPU based MPS produce better performance compared to the traditional arrangement of using CPUs.

NnAudio: An on-the-Fly GPU Audio to Spectrogram Conversion Toolbox Using 1D Convolutional Neural Networks

In this paper, we present nnAudio, a new neural network-based audio processing framework with graphics processing unit (GPU) support that leverages 1D convolutional neural networks to perform time domain to frequency domain conversion. It allows on-the-fly spectrogram extraction due to its fast speed, without the need to store any spectrograms on the disk. Moreover, this approach also allows back-propagation on the waveforms-to-spectrograms transformation layer, and hence, the transformation process can be made trainable, further optimizing the waveform-to-spectrogram transformation for the specific task that the neural network is trained on. All spectrogram implementations scale as Big-O of linear time with respect to the input length. nnAudio, however, leverages the compute unified device architecture (CUDA) of 1D convolutional neural network from PyTorch, its short-time Fourier transform (STFT), Mel spectrogram, and constant-Q transform (CQT) implementations are an order of magnitude faster than other implementations using only the central processing unit (CPU). We tested our framework on three different machines with NVIDIA GPUs, and our framework significantly reduces the spectrogram extraction time from the order of seconds (using a popular python library librosa) to the order of milliseconds, given that the audio recordings are of the same length. When applying nnAudio to variable input audio lengths, an average of 11.5 hours are required to extract 34 spectrogram types with different parameters from the MusicNet dataset using librosa. An average of 2.8 hours is required for nnAudio, which is still four times faster than librosa. Our proposed framework also outperforms existing GPU processing libraries such as Kapre and torchaudio in terms of processing speed.

Improving classification and clustering techniques using GPUs

SummaryClassification and clustering techniques are used in different applications. Large‐scale big data applications such as social networks analysis applications need to process large data chunks in a short time. Classification and clustering tasks in such applications consume a lot of processing time. Improving the performance of classification and clustering algorithms enhances the performance of applications that use such type of algorithms. This paper introduces an approach for exploiting the graphics processing unit (GPU) platform to improve the performance of classification and clustering algorithms. The proposed approach uses two GPUs implementations, which are the pure GPU or GPU‐only implementation and the GPU‐CPU hybrid implementation. The results show that the hybrid implementation, which optimizes the subtask scheduling for both the CPU and the GPU processing elements, outperforms the approach that uses only the GPU.

Accelerated organ region segmentation by the revised radial basis function network using a graphics processing unit.

This study aimed to accelerate the segmentation of organs in medical imaging with the revised radial basis function (RBF) network, using a graphics processing unit (GPU). We segmented the lung and liver regions from 250 chest x-ray computed tomography (CT) images and 160 abdominal CT images, respectively, using the revised RBF network. We compared the time taken to segment images and their accuracy between serial processing by a single-core central processing unit (CPU), parallel processing using four CPU cores, and GPU processing. Segmentation times for lung and liver organ regions shortened to 57.80 and 35.35 seconds for CPU parallel processing and 20.16 and 11.02 seconds for GPU processing, compared to 211.03 and 124.21 seconds for CPU serial processing, respectively. The concordance rate of the segmented region to the normal region in slices excluding the upper and lower ends (173 lung and 111 liver slices) was 98% for lung and 96% for liver. The use of CPU parallel processing and GPU shortened the organ segmentation time in the revised RBF network without compromising segmentation accuracy. In particular, segmentation time was shortened to less than 10%with GPU. This processing method will contribute to workload reduction in imaging analysis. J. Med. Invest. 66 : 86-92, February, 2019.

Layer-based visualization and biomedical information exploration of multi-channel large histological data

Background and objectiveModern microscopes can acquire multi-channel large histological data from tissues of human beings or animals, which contain rich biomedical information for disease diagnosis and biological feature analysis. However, due to the large size, fuzzy tissue structure, and complicated multiple elements integrated in the image color space, it is still a challenge for current software systems to effectively calculate histological data, show the inner tissue structures and unveil hidden biomedical information. Therefore, we developed new algorithms and a software platform to address this issue. MethodsThis paper presents a multi-channel biomedical data computing and visualization system that can efficiently process large 3D histological images acquired from high-resolution microscopes. A novelty of our system is that it can dynamically display a volume of interest and extract tissue information using a layer-based data navigation scheme. During the data exploring process, the actual resolution of the loaded data can be dynamically determined and updated, and data rendering is synchronized in four display windows at each data layer, where 2D textures are extracted from the imaging volume and mapped onto the displayed clipping planes in 3D space. ResultsTo test the efficiency and scalability of this system, we performed extensive evaluations using several different hardware systems and large histological color datasets acquired from a CryoViz 3D digital system. The experimental results demonstrated that our system can deliver interactive data navigation speed and display detailed imaging information in real time, which is beyond the capability of commonly available biomedical data exploration software platforms. ConclusionTaking advantage of both CPU (central processing unit) main memory and GPU (graphics processing unit) graphics memory, the presented software platform can efficiently compute, process and visualize very large biomedical data and enhance data information. The performance of this system can satisfactorily address the challenges of navigating and interrogating volumetric multi-spectral large histological image at multiple resolution levels.

Live Ultrasound Color-Encoded Speckle Imaging Platform for Real-Time Complex Flow Visualization In Vivo.

Complex flow patterns are prevalent in the vasculature, but they are difficult to image noninvasively in real time. This paper presents the first real-time scanning platform for a high-frame-rate ultrasound technique called color-encoded speckle imaging (CESI) and its use in visualizing arterial flow dynamics in vivo. CESI works by simultaneously rendering flow speckles and color-coded flow velocity estimates on a time-resolved basis. Its live implementation was achieved by integrating a 192-channel programmable ultrasound front-end module, a 4.8-GB/s capacity data streaming link, and a series of computing kernels implemented on the graphical processing unit (GPU) for beamforming and Doppler processing. A slow-motion replay mode was also included to offer coherent visualization of CESI frames acquired at high frame rate [3000 frames per second (fps) in our experiments]. The live CESI scanning platform was found to be effective in facilitating real-time image guidance (at least 20 fps for live video display with 55-fps GPU processing throughout). In vivo pilot trials also showed that live CESI, when running in replay mode, can temporally resolve triphasic flow at the brachial bifurcation and can reveal flow dynamics in the brachial vein during a fist-clenching maneuver. Overall, live CESI has potential for use in routine investigations in vivo that seek to identify complex flow dynamics in real time and relate these dynamics to vascular physiology.