Graphics Processing Units Research Articles

NGS: A network GPGPU system for orchestrating remote and virtual accelerators

In General-Purpose computing on Graphics Processing Unit (GPGPU), the use of CPUs is combined with that of GPUs. CPUs are used for sequential code, while GPUs are used for parallel code. GPGPU has been enabled by two key factors: (i) the massively parallel architecture of GPUs, which allows thousands of single cores to run parallel code; and (ii) the development of platforms, such as CUDA, that simplify implementing code for GPUs. GPGPU has established itself as the standard computing system in most computing fields due to the great improvements it brings. However, its use is not without problems, such as GPU underutilization, high cost, power consumption, etc. In this paper we present NGS (Network GPGPU System) to address the underutilization of GPUs in computing centers. NGS orchestrates the concurrent access to GPGPU resources from different nodes of the cluster by leveraging the remote GPU virtualization mechanism and the NVML library by NVIDIA. In this way, NGS enables different nodes of the cluster to access remote GPUs as if they were local at the same time that this access is guaranteed to be carried out without collisions. The main novelty is that NGS offers a global and standard solution independent of the computing environment used. Experimental results show up to 4x improvements compared to popular approaches.

A parallel and performance portable implementation of a full-field crystal plasticity model

We have developed a parallel implementation of an Elasto-Viscoplastic Fast Fourier Transform-based (EVPFFT) micromechanical solver to enable computationally efficient crystal plasticity modeling for polycrystalline materials. Our primary focus lies in achieving performance portability, allowing a single EVPFFT implementation to run optimally on various homogeneous architectures, including multi-core Central Processing Units (CPUs), as well as on heterogeneous computer architectures comprising multi-core CPUs and Graphics Processing Units (GPUs) from different vendors. To accomplish this goal, we have leveraged MATAR, a C++ software library that simplifies the creation and utilization of multidimensional dense or sparse matrix and array data structures. These data structures are designed to be portable across diverse architectures through the use of Kokkos, a performance-portable library. Additionally, we have employed the Message Passing Interface (MPI) to efficiently distribute the computational workload among processors. The heFFTe (Highly Efficient FFT for Exascale) library is used to facilitate the performance portability of the fast Fourier transforms (FFTs) computation. The computational performance of EVPFFT is evaluated and presented in terms of parallel scalability and simulation runtime on different high-performance computing (HPC) architectures. The utility of the developed framework to efficiently simulate the micro-mechanical fields in polycrystalline microstructures in engineering applications is discussed. Program summaryProgram Title: EVPFFTCPC Library link to program files:https://doi.org/10.17632/2k8579fyyv.1Developer's repository link:https://github.com/lanl/FierroLicensing provisions: BSD 3-Clause LicenseProgramming language: C++External routines/libraries: MPI, Kokkos, MATAR, HeFFTe, HDF5Nature of problem: EVPFFT is a crystal plasticity code designed to compute micro-mechanical fields within a polycrystalline representative volume element (RVE) and predict the macroscale response of the RVE.Solution method: EVPFFT uses the periodic Green's function method in Fourier space to solve the field equations of static stress equilibrium in a periodic spatial domain.

A robust neural network for privacy-preserving heart rate estimation in remote healthcare systems

In this study, we propose a computationally-light and robust neural network for estimating heart rate in remote healthcare systems. We develop a model that can be trained on consumer-grade graphics processing units (GPUs), and can be deployed on edge devices for swift inference. We propose a hybrid model based on convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM) architectures for estimating heart rate from Electrocardiogram (ECG) and Photoplethysmography (PPG) signals. Considering the sensitive nature of the ECG signals, we ensure a formal privacy guarantee, differential privacy, for the model training. We perform a tight accounting of the overall privacy budget of our training algorithm using the Rényi Differential Privacy technique. We demonstrate that our model outperforms state-of-the-art networks on a benchmark dataset for both ECG and PPG signals despite having a much smaller number of trainable parameters and, consequently, much smaller training and inference times. Our CNN-BiLSTM architecture can also provide excellent heart rate estimation performance even under strict privacy constraints. We develop a prototype Arduino-based data collection system that is low-cost, efficient, and useful for providing access to modern healthcare services to people living in remote areas.

HD-TMA: A New Fast Template Matching Algorithm Implementation for Linear DAS Array Data and Its Optimization Strategies

Abstract Distributed acoustic sensing (DAS) technology, combined with existing telecom fiber-optic cable, has shown great potential in earthquake monitoring. The template matching algorithm (TMA) shows good detection capabilities but depends on heavy computational cost and diverse template events. We developed a program named HD-TMA (high-efficiency DAS template matching algorithm), which accelerates computation by 40 times on the central processing unit platform and 2 times on the graphic processing unit platform. For linear DAS array data, we introduced a fast arrival-picking algorithm based on the Hough transform to pick the time window of template waveform. The HD-TMA was successfully applied to the 2022 Ms 6.9 Menyuan earthquake aftershock sequence recorded by a DAS array, and the DAS data result was compared with a collocated short-period seismometer data’s result. Two optimization strategies were discussed based on this data set. (1) Using signal-to-noise ratio in choosing the location and aperture of the subarray and the time window of the template waveform. (2) Considering the decrease in template events’ marginal utility, we proposed applying a neural network to build a template event library, followed by the HD-TMA scanning. Such strategies can effectively reduce computational cost and improve detection capability.

Analyzing the performance of a bio-sensor integrated improved blended learning model for accurate pneumonia prediction

Pneumonia has been considered a life-threatening disease for elderly human beings and those with weakened immune systems in the present medical era. The contemporary scenario highlights the significance of intelligent and automatic handheld devices to detect pneumonia and other pulmonary diseases. Hence, this research has designed an improved blended learning paradigm (IBLP) for real-time pneumonia detection from chest X-rays, and early detection of lung diseases from alveolar gas using biosensors with a graphical processing unit (GPU) developed to overcome and resolve such challenges. It emphasizes the medical applications of blended learning techniques, particularly for identifying pneumonia from chest X-ray images and other pulmonary diseases from exhaled breath using biosensors and support vector machine (SVM). The experimental findings indicate that the blended learning based VGG16 (91.99%) consistently outperforms the VGG19 (88.91%) and ResNet50 (87.02%) model in diagnostic accuracy. IBLP VGG16 provided 95.5% precision, 97.69% F1 score, and a 100% recall rate with no false-negative results. The future of X-ray classification in pneumonia diagnosis will likely involve using artificial intelligence-based systems that can provide accurate and timely analysis of X-ray images, thereby improving patient outcomes and reducing healthcare costs.

Hand-held GPU accelerated device for multiclass classification of X-ray images using CNN model

Chest X-ray (CXR) images are the primary investigation aid for many lung diseases and their follow-ups. For diagnosis of SARS-CoV-2, RT–PCR test and chest Computed Tomography (CT) are commonly used but both face false negatives for ruling out the infection. So, there is a demanding need for developing a system combined with Artificial Intelligence (AI) and CXR imaging to detect COVID-19 patients to avoid its spread. Here, a robust and efficient handheld device is proposed. It uses the computational power of the Graphics Processing Unit (GPU) and pre-trained deep learning models for analyzing the CXR images. A Resnet-50 CNN model is deployed on an NVIDIA Jetson Nano GPU module for the real-time classification of COVID-19, Tuberculosis, and Normal using CXR images. The device can perform real-time classification of CXR images from a portable X-ray machine and classify the image into one of the above categories. For the extensive training, a database of 680 COVID-19, 1230 Tuberculosis, and 1050 normal CXR images are extracted by combining several global databases like Kaggle, SIRM, RSNA, and Radiopaedia. The classification accuracy, precision, and loss rate were 0.9879, 0.9758, and 0.0196 respectively and our model would improve with larger data sets. The highly accurate and high-performance GPU device significantly plays a far-reaching role in COVID-19 diagnosis using Chest X-ray, which could be beneficial to triage the health system and to combat the outbreak of COVID-19.

Graphics processing unit (GPU)-enhanced nonhydrostatic model with grid nesting for global tsunami propagation and coastal inundation

Nonhydrostatic models have proven their superiority in describing tsunami propagation over trans-oceanic distances and nearshore transformation because of their good dispersion and nonlinearity properties. The novel one-layer nonhydrostatic formulations proposed by Wang et al. [Phys. Fluids 35, 076610 (2023)] have been rederived in the spherical coordinate system incorporating Coriolis effects to enable the application of basin-wide tsunami propagation. The model was implemented using the fractional step method, where the hydrostatic step was solved by a Godunov-type finite-volume scheme, and the nonhydrostatic step was obtained with the finite-difference method. Additionally, a two-way grid-nesting scheme was employed to adapt the topographic features for efficient computation of tsunami propagation in deep ocean and coastal inundation. Furthermore, graphics processing unit (GPU)-parallelism technique was incorporated to further optimize the model performance. An idealized benchmark test as well as three experiments of regular and irregular waves, solitary, and N-waves transformations have been simulated to demonstrate the superior performance of the current GPU-accelerated grid-nesting nonhydrostatic model. Finally, the model has been applied to reproduce the 1964 Prince William Sound Tsunami, its propagation across the North Pacific and induced inundation in the Seaside.

Building Depth Maps Using an Active-Pulse Television Measuring System in Real Time Domain

The paper presents the results of software development for building depth maps based on video data from a television camera of an active-pulse television measuring system (AP TMS) in real time domain. The development of such software is required to conduct various scientific studies, as well as to improve the methods and techniques for building depth maps and remote measurement of the characteristics of objects of interest. The software was implemented using the Python programming language with additional libraries installed. According to the results of testing the implemented algorithm, it was found that the calculation speed using the graphics processing unit (GPU) is on average 3.5 times higher than the speed of the algorithm using only the central processing unit (CPU). It has been established that with the help of CUDA cores it is possible to build depth maps in real time domain at the maximum possible resolution of video frames of the system (1544x2064 pixels), while when using the central processor, real-time operation is possible only at a reduced resolution of video frames (772x1032 pixels).

Investigation of friction modeling on numerical Ti6Al4V cutting simulations

Friction significantly influences chip formation, thereby highlighting its modeling critical in numerical cutting simulations. Notably, issues like the underestimation of feed force in simulations are often attributed to inadequate friction models. Nevertheless, diverse conclusions in the literature regarding friction’s behavior complicate the accurate implementation of this input. Additionally, the enormous number of the available friction models and their varied calibration methods introduce further debates over which aspects of friction modeling should receive more focus. These controversies and deficiencies hinder progress in further understanding friction’s role in chip formation through numerical studies. Instead of proposing or calibrating new friction models, the current study, based on Ti6Al4V machining simulations, attempts to readdress the aforementioned controversies in a neutral stance by conducting sensitivity studies using a hybrid Smoothed Particle Hydrodynamics (SPH) - Finite Element Method (FEM) solver with Graphics Processing Units (GPU) acceleration, which is capable of efficiently executing high-resolution computations. The significant impact of friction particularly at the end of the tool-chip contact on the chip formation, physical contact states and process forces are highlighted. The behaviors of physical parameters including normal contact pressure, sliding velocity, and temperature-dependent friction models in the literature are also evaluated. Several aspects such as the shear flow stress limit, the relationship between friction and process forces, and the selection of different models are discussed. In conclusion, pragmatic recommendations for friction modeling and cutting simulation work are provided. On the one hand, the available complex physical parameter dependent friction models could not prove their necessity and should be approached cautiously. Instead, the constant Coulomb friction model without the shear stress limit, despite its simplicity, demonstrates effective and sufficient for a single set of metal cutting simulations. On the other hand, reliable on-site measurement techniques of the coefficient of friction (COF) at the tool-chip sliding contact area should be developed, with consideration of the contact length and the state of material flow. Combined with careful cutting edge preparation and suitable constitutive models, the overall accuracy of numerical cutting simulations including the feed force prediction is expected to be improved.

Plant Leaf Disease Detection Using CNN

The existence of pests and diseases in plants and crops has a substantial impact on agricultural production within a country. Monitoring plants meticulously to detect and identify diseases is a common practice among farmers and experts. However, this approach is often laborious, costly, and not entirely reliable. To mitigate this issue, we propose a Disease Recognition Model based on leaf image classification. Our objective is to detect plant diseases using image processing techniques, specifically leveraging Convolutional Neural Networks (CNNs). CNNs are a category of artificial neural networks tailored for handling pixel-based inputs, particularly adept at image recognition tasks. Abbreviation- CNN (Convolutional Neural Network), SVM (Support Vector Machine), KNN (K Nearest Neighbour), ANN (Artificial Neural Network), GPU (Graphics processing unit)

Time Predictable Modeling Method for GPU Architecture with SIMT and Cache Miss Awareness

Graphics Processing Units (GPUs) are used to accelerate computing-intensive tasks, such as neural networks,data analysis, high-performance computing, etc. In the past decade or so, researchers have done a lot of work on GPU architecture and proposed a variety of theories and methods to study the microarchitectural characteristics of various GPUs. In this study, the GPU serves as a co-processor and works together with the CPU in an embedded real-time system to handle computationally intensive tasks. It models the architecture of the GPU and further considers it based on some excellent work. The SIMT mechanism and Cache-miss situation provide a more detailed analysis of the GPU architecture. In order to verify the GPU architecture model proposed in this article, 10 GPU kernel_task and an Nvidia GPU device were used to perform experiments. The experimental results showed that the minimum error between the kernel task execution time predicted by the GPU architecture model proposed in this article and the actual measured kernel task execution time was 3.80%, and the maximum error was 8.30%.

Localization of overdamped bosonic modes and transport in strange metals

The strange metal phase of correlated electrons materials was described in a recent theory by a model of a Fermi surface coupled a two-dimensional quantum critical bosonic field with a spatially random Yukawa coupling. With the assumption of self-averaging randomness, similar to that in the Sachdev-Ye-Kitaev model, numerous observed properties of a strange metal were obtained for a wide range of intermediate temperatures, including the linear in temperature resistivity. The Harris criterion implies that spatial fluctuations in the local position of the critical point must dominate at lower temperatures. For an [Formula: see text]-component boson with [Formula: see text], we use multiple graphics processing units (GPUs) to compute the real frequency spectrum of the boson propagator in a self-consistent mean-field treatment of the boson self-interactions, but an exact treatment of multiple realizations of the spatial randomness from the random boson mass. We find that Landau damping from the fermions leads to the emergence of the physics of the random transverse-field Ising model at low temperatures, as has been proposed by Hoyos, Kotabage, and Vojta. This regime is controlled by localized overdamped eigenmodes of the bosonic scalar field, also has a resistivity which is nearly linear-in-temperature, and extends into a "quantum critical phase" away from the quantum critical point, as observed in several cuprates. For the [Formula: see text] Ising scalar, the mean-field treatment is not applicable, and so we use Hybrid Monte Carlo simulations running on multiple GPUs; we find a rounded transition and localization physics, with strange metal behavior in an extended region around the transition.

SSA2py: A High-Performance Python Implementation of the Source-Scanning Algorithm for Spatiotemporal Seismic Source Imaging

Abstract This article introduces the first version of SSA2py (v.1.0)—an open-source package designed to implement the source-scanning algorithm (SSA). SSA2py is a Python-based, high-performance-oriented package that incorporates the SSA method, which has been effectively applied to numerous earthquakes for imaging the spatiotemporal behavior of the seismic source. The software supports a wide range of data and metadata resources. These include the International Federation of Digital Seismograph Networks Web Services, the SeedLink protocol, and others, ensuring optimal access to waveforms and station metadata. Furthermore, the code may evaluate the quality of accessible waveforms using signal analysis methods, allowing for the most appropriate data selection. The SSA method has been computationally optimized using multiprocessing techniques for efficient central processing unit and graphic processing units executions, enabling considerably accelerated computational processes even for large-scale grid searches. The program is also designed to provide statistical and methodological uncertainties for the executed cases through jackknife, bootstrap, and backprojection array response function tests. After appropriate tuning by the user, SSA2py can be used for detailed earthquake source studies that backprojection technique typically serves as a complementary output to the source inversion result or as a near-real-time tool for successful and quick identification of the style and complexity of the earthquake rupture. With a wide and flexible configuration, the user has complete control over all calculating aspects of SSA2py. This article provides a detailed description of the structure and capabilities of this new package, and its reliability is demonstrated through targeted applications to the 2004 Mw 6.0 Parkfield and 2019 Mw 7.1 Ridgecrest earthquakes. Furthermore, the computational efficiency of SSA2py is validated through rigorous performance tests.

ACCELERATING ORE SINTERING MATHEMATICAL MODEL USING GPU

The study aims to enhance the efficiency and computational speed of the ore sintering model through the utilization of graphics processing units (GPUs). The purpose of this research is to address the growing demand for faster and more scalable simulations in the field of ore sintering, a crucial process in the production of iron and steel. Methodology involves the integration of parallel computing capabilities offered by GPUs into the existing ore sintering model. By leveraging the parallel processing power of GPUs, the computational workload is distributed across multiple cores, significantly reducing the simulation time. Results demonstrate a substantial acceleration in the ore sintering simulation process. Comparative analyses between CPU and GPU implementations reveal a remarkable reduction in computation time, thereby enabling real-time or near-real-time simulations. The achieved speedup not only enhances the efficiency of ore sintering modeling but also opens avenues for exploring larger and more complex scenarios. This is the successful integration of GPU parallel computing into the ore sintering model, showcasing the adaptability of advanced computational technologies to traditional industrial processes. The study contributes to the field by bridging the gap between computational power and metallurgical simulations, demonstrating the potential for GPU acceleration in other areas of metallurgical processes. Practical significance of this research is underscored by its potential to revolutionize the ore sintering industry. Faster simulations facilitate quicker decision-making in process optimization, leading to improved energy efficiency and reduced environmental impact. This research sets the stage for the broader adoption of GPU acceleration in metallurgical modeling, signaling a paradigm shift towards more efficient and sustainable industrial practices

Fast knowledge graph completion using graphics processing units

Knowledge graphs can be used in many areas related to data semantics such as question-answering systems, knowledge based systems. However, the currently constructed knowledge graphs need to be complemented for better knowledge in terms of relations. It is called knowledge graph completion. To add new relations to the existing knowledge graph by using knowledge graph embedding models, we have to evaluate N×N×R vector operations, where N is the number of entities and R is the number of relation types. It is very costly.In this paper, we provide an efficient knowledge graph completion framework on GPUs to get new relations using knowledge graph embedding vectors. In the proposed framework, we first define transformable to a metric space and then provide a method to transform the knowledge graph completion problem into the similarity join problem for a model which is transformable to a metric space. After that, to efficiently process the similarity join problem, we derive formulas using the properties of a metric space. Based on the formulas, we develop a fast knowledge graph completion algorithm. Finally, we experimentally show that our framework can efficiently process the knowledge graph completion problem.

An image-based metal artifact reduction technique utilizing forward projection in computed tomography.

The projection data generated via the forward projection of a computed tomography (CT) image (FP-data) have useful potentials in cases where only image data are available. However, there is a question of whether the FP-data generated from an image severely corrupted by metal artifacts can be used for the metal artifact reduction (MAR). The aim of this study was to investigate the feasibility of a MAR technique using FP-data by comparing its performance with that of a conventional robust MAR using projection data normalization (NMARconv). The NMARconv was modified to make use of FP-data (FPNMAR). A graphics processing unit was used to reduce the time required to generate FP-data and subsequent processes. The performances of FPNMAR and NMARconv were quantitatively compared using a normalized artifact index (AIn) for two cases each of hip prosthesis and dental fillings. Several clinical CT images with metal artifacts were processed by FPNMAR. The AIn values of FPNMAR and NMARconv were not significantly different from each other, showing almost the same performance between these two techniques. For all the clinical cases tested, FPNMAR significantly reduced the metal artifacts; thereby, the images of the soft tissues and bones obscured by the artifacts were notably recovered. The computation time per image was ~ 56ms. FPNMAR, which can be applied to CT images without accessing the projection data, exhibited almost the same performance as that of NMARconv, while consuming significantly shorter processing time. This capability testifies the potential of FPNMAR for wider use in clinical settings.

Advanced Framework for Multi-Modal Healthcare Data Integration: Leveraging HPC with GPU Computing and CNN Architecture in CDSS

In this study, we shall be looking at the challenges involved in integrating multi-modal healthcare data in the clinical decision support systems (CDSS). We propose the Automated Multi-Modal Data Integration (AMMI-CDSS) algorithm, which will utilize the latest high-performance computing (HPC) techniques such as the Convolutional Neural Network (CNN) architecture and the Graphics Processing Unit (GPU) computing to provide precise and rapid analysis. Which features will be extracted, multi-modal data will be merged, data will be prepared and algorithms developed in a distributed computing environment. We illustrate how AMMI-CDSS through the use of real world datasets such as wearable sensors data, medical imaging, genetic data, and electronic health records (EHRs), can improve the clinical decision support. By performing harmonization of the diverse data sources into a unique dataset after thorough data preprocessing and complex calculations, AMMI-CDSS provides the analysis with better quality and coherence.
 Our study allow us to make conclusion about how HPC-based CDSS models can be compared to conventional machine learning ones using their scalability and performance as key metrics. We enrich CDSS with the methodical framework for one-by-one testing and evaluation of proposed models and multi-modal healthcare data analysis. Future research might explore novel methods for integrating diverse types of healthcare data, as well as enhancing the HPC-based CDSS models by keeping them up-to-date.

Accelerating flash flood simulations: An efficient GPU implementation for a slim shallow water solver

Flash floods are an increasing hazard to human infrastructure and life. Effective disaster management and mitigation require accurate and fast predictions for decision-making. The Flood Inundation Parallel Computation (FIP) software presented in this paper shows how to fully exploit the tremendous computational capabilities of graphics processing units (GPUs) to accelerate shallow water solvers beyond the current state-of-the-art. The time efficient explicit shallow water scheme for structured grids RMG, is introduced and implemented in FIP. The optimized GPU implementation of FIP achieves balanced memory- and instruction throughputs of up to 80%. Validation and performance tests using laboratory and historical flood cases are presented that demonstrate the accuracy and effectiveness of FIP in predicting flood inundation. Our implementation achieves a fourfold speedup in comparison to state-of-the-art approaches. It enables faster-than real-time simulations of areas of over 600 km2 at 1 m resolution on consumer-grade GPUs.

Time-domain event detection using single-instruction, multiple-thread gpGPU architectures in single-molecule biophysical data

Discrete amplitude levels in ordered, time-domain data often represent different underlying latent states of the system that is being interrogated. Analysis and feature extraction from these data sets generally requires considering the order of each individual point; this approach cannot take advantage of contemporary general-purpose graphics processing units (gpGPU) and single-instruction multiple-data (SIMD) instruction set architectures. Two sources of such data from single-molecule biological measurements are nanopores and single-molecule field effect transistor (smFET) nanotube devices; both generate streams of time-ordered current or voltage data, typically sampled near 1 MS/s, with run times of minutes, yielding terabyte-scale datasets. Here, we present three gpGPU-based algorithms to overcome limitations associated with serial event detection in time series data, resulting in a 250× improvement in the rate with which we can detect salient features in nanopore and smFET datasets. The code is freely available.

Robust image descriptor for machine learning based data reduction in serial crystallography.

Serial crystallography experiments at synchrotron and X-ray free-electron laser (XFEL) sources are producing crystallographic data sets of ever-increasing volume. While these experiments have large data sets and high-frame-rate detectors (around 3520 frames per second), only a small percentage of the data are useful for downstream analysis. Thus, an efficient and real-time data classification pipeline is essential to differentiate reliably between useful and non-useful images, typically known as 'hit' and 'miss', respectively, and keep only hit images on disk for further analysis such as peak finding and indexing. While feature-point extraction is a key component of modern approaches to image classification, existing approaches require computationally expensive patch preprocessing to handle perspective distortion. This paper proposes a pipeline to categorize the data, consisting of a real-time feature extraction algorithm called modified and parallelized FAST (MP-FAST), an image descriptor and a machine learning classifier. For parallelizing the primary operations of the proposed pipeline, central processing units, graphics processing units and field-programmable gate arrays are implemented and their performances compared. Finally, MP-FAST-based image classification is evaluated using a multi-layer perceptron on various data sets, including both synthetic and experimental data. This approach demonstrates superior performance compared with other feature extractors and classifiers.