Search
Search results for: hybrid cpu+gpu system
-
Efficient parallel implementation of crowd simulation using a hybrid CPU+GPU high performance computing system
PublicationIn the paper we present a modern efficient parallel OpenMP+CUDA implementation of crowd simulation for hybrid CPU+GPU systems and demonstrate its higher performance over CPU-only and GPU-only implementations for several problem sizes including 10 000, 50 000, 100 000, 500 000 and 1 000 000 agents. We show how performance varies for various tile sizes and what CPU–GPU load balancing settings shall be preferred for various domain...
-
Parallelization of large vector similarity computations in a hybrid CPU+GPU environment
PublicationThe paper presents design, implementation and tuning of a hybrid parallel OpenMP+CUDA code for computation of similarity between pairs of a large number of multidimensional vectors. The problem has a wide range of applications, and consequently its optimization is of high importance, especially on currently widespread hybrid CPU+GPU systems targeted in the paper. The following are presented and tested for computation of all vector...
-
Investigation of Parallel Data Processing Using Hybrid High Performance CPU + GPU Systems and CUDA Streams
PublicationThe paper investigates parallel data processing in a hybrid CPU+GPU(s) system using multiple CUDA streams for overlapping communication and computations. This is crucial for efficient processing of data, in particular incoming data stream processing that would naturally be forwarded using multiple CUDA streams to GPUs. Performance is evaluated for various compute time to host-device communication time ratios, numbers of CUDA streams,...
-
Implementation of FDTD-compatible Green's function on heterogeneous CPU-GPU parallel processing system
PublicationThis paper presents an implementation of the FDTD-compatible Green's function on a heterogeneous parallel processing system. The developed implementation simultaneously utilizes computational power of the central processing unit (CPU) and the graphics processing unit (GPU) to the computational tasks best suited to each architecture. Recently, closed-form expression for this discrete Green's function (DGF) was derived, which facilitates...
-
Auto-tuning methodology for configuration and application parameters of hybrid CPU + GPU parallel systems based on expert knowledge
PublicationAuto-tuning of configuration and application param- eters allows to achieve significant performance gains in many contemporary compute-intensive applications. Feasible search spaces of parameters tend to become too big to allow for exhaustive search in the auto-tuning process. Expert knowledge about the utilized computing systems becomes useful to prune the search space and new methodologies are needed in the face of emerging heterogeneous...
-
Tuning a Hybrid GPU-CPU V-Cycle Multilevel Preconditioner for Solving Large Real and Complex Systems of FEM Equations
PublicationThis letter presents techniques for tuning an accelerated preconditioned conjugate gradient solver with a multilevel preconditioner. The solver is optimized for a fast solution of sparse systems of equations arising in computational electromagnetics in a finite element method using higher-order elements. The goal of the tuning is to increase the throughput while at the same time reducing the memory requirements in order to allow...
-
Parallelization of Selected Algorithms on Multi-core CPUs, a Cluster and in a Hybrid CPU+Xeon Phi Environment
PublicationIn the paper we present parallel implementations as well as execution times and speed-ups of three different algorithms run in various environments such as on a workstation with multi-core CPUs and a cluster. The parallel codes, implementing the master-slave model in C+MPI, differ in computation to communication ratios. The considered problems include: a genetic algorithm with various ratios of master processing time to communication...
-
Acceleration of the DGF-FDTD method on GPU using the CUDA technology
PublicationWe present a parallel implementation of the discrete Green's function formulation of the finite-difference time-domain (DGF-FDTD) method on a graphics processing unit (GPU). The compute unified device architecture (CUDA) parallel computing platform is applied in the developed implementation. For the sake of example, arrays of Yagi-Uda antennas were simulated with the use of DGF-FDTD on GPU. The efficiency of parallel computations...
-
Characterizing the Scalability of Graph Convolutional Networks on Intel® PIUMA
PublicationLarge-scale Graph Convolutional Network (GCN) inference on traditional CPU/GPU systems is challenging due to a large memory footprint, sparse computational patterns, and irregular memory accesses with poor locality. Intel’s Programmable Integrated Unffied Memory Architecture (PIUMA) is designed to address these challenges for graph analytics. In this paper, a detailed characterization of GCNs is presented using the Open-Graph Benchmark...
-
Finite element matrix generation on a GPU
PublicationThis paper presents an efficient technique for fast generation of sparse systems of linear equations arising in computational electromagnetics in a finite element method using higher order elements. The proposed approach employs a graphics processing unit (GPU) for both numerical integration and matrix assembly. The performance results obtained on a test platform consisting of a Fermi GPU (1x Tesla C2075) and a CPU (2x twelve-core...
-
Jacobi and gauss-seidel preconditioned complex conjugate gradient method with GPU acceleration for finite element method
PublicationIn this paper two implementations of iterative solvers for solving complex symmetric and sparse systems resulting from finite element method applied to wave equation are discussed. The problem under investigation is a dielectric resonator antenna (DRA) discretized by FEM with vector elements of the second order (LT/QN). The solvers use the preconditioned conjugate gradient (pcg) method implemented on Graphics Processing Unit (GPU)...
-
Benchmarking Performance of a Hybrid Intel Xeon/Xeon Phi System for Parallel Computation of Similarity Measures Between Large Vectors
PublicationThe paper deals with parallelization of computing similarity measures between large vectors. Such computations are important components within many applications and consequently are of high importance. Rather than focusing on optimization of the algorithm itself, assuming specific measures, the paper assumes a general scheme for finding similarity measures for all pairs of vectors and investigates optimizations for scalability...
-
Block Conjugate Gradient Method with Multilevel Preconditioning and GPU Acceleration for FEM Problems in Electromagnetics
PublicationIn this paper a GPU-accelerated block conjugate gradient solver with multilevel preconditioning is presented for solving large system of sparse equations with multiple right hand-sides (RHSs) which arise in the finite-element analysis of electromagnetic problems. We demonstrate that blocking reduces the time to solution significantly and allows for better utilization of the computing power of GPUs, especially when the system matrix...
-
Parallel implementation of the DGF-FDTD method on GPU Using the CUDA technology
PublicationThe discrete Green's function (DGF) formulation of the finite-difference time-domain method (FDTD) is accelerated on a graphics processing unit (GPU) by means of the Compute Unified Device Architecture (CUDA) technology. In the developed implementation of the DGF-FDTD method, a new analytic expression for dyadic DGF derived based on scalar DGF is employed in computations. The DGF-FDTD method on GPU returns solutions that are compatible...
-
Single and Dual-GPU Generalized Sparse Eigenvalue Solvers for Finding a Few Low-Order Resonances of a Microwave Cavity Using the Finite-Element Method
PublicationThis paper presents two fast generalized eigenvalue solvers for sparse symmetric matrices that arise when electromagnetic cavity resonances are investigated using the higher-order finite element method (FEM). To find a few loworder resonances, the locally optimal block preconditioned conjugate gradient (LOBPCG) algorithm with null-space deflation is applied. The computations are expedited by using one or two graphical processing...
-
Performance and Energy Aware Training of a Deep Neural Network in a Multi-GPU Environment with Power Capping
PublicationIn this paper we demonstrate that it is possible to obtain considerable improvement of performance and energy aware metrics for training of deep neural networks using a modern parallel multi-GPU system, by enforcing selected, non-default power caps on the GPUs. We measure the power and energy consumption of the whole node using a professional, certified hardware power meter. For a high performance workstation with 8 GPUs, we were...
-
A GPU Solver for Sparse Generalized Eigenvalue Problems with Symmetric Complex-Valued Matrices Obtained Using Higher-Order FEM
PublicationThe paper discusses a fast implementation of the stabilized locally optimal block preconditioned conjugate gradient (sLOBPCG) method, using a hierarchical multilevel preconditioner to solve nonHermitian sparse generalized eigenvalue problems with large symmetric complex-valued matrices obtained using the higher-order finite-element method (FEM), applied to the analysis of a microwave resonator. The resonant frequencies of the low-order...
-
Implementation of TVDI calculation for coastal zone
PublicationPaper will show an implementation of TVDI (Temperature-Vegetation-Dryness Index) algorithm on GPU (Graphics Processing Unit). Calculation of this index is based on LST (Land Surface Temperature) and NDVI (Normalized Difference Vegetation Index). Discussed results are based on multi-spectral imagery retrieved from AVHRR3 sensors for area of Poland, especially from region of Gdańsk coastal zone. All phases of TVDI implementation...
-
Optimization of Data Assignment for Parallel Processing in a Hybrid Heterogeneous Environment Using Integer Linear Programming
PublicationIn the paper we investigate a practical approach to application of integer linear programming for optimization of data assignment to compute units in a multi-level heterogeneous environment with various compute devices, including CPUs, GPUs and Intel Xeon Phis. The model considers an application that processes a large number of data chunks in parallel on various compute units and takes into account computations, communication including...
-
GPU-Accelerated Finite-Element Matrix Generation for Lossless, Lossy, and Tensor Media [EM Programmer's Notebook]
PublicationThis paper presents an optimization approach for limiting memory requirements and enhancing the performance of GPU-accelerated finite-element matrix generation applied in the implementation of the higher-order finite-element method (FEM). It emphasizes the details of the implementation of the matrix-generation algorithm for the simulation of electromagnetic wave propagation in lossless, lossy, and tensor media. Moreover, the impact...
-
Generation of large finite-element matrices on multiple graphics processors
PublicationThis paper presents techniques for generating very large finite-element matrices on a multicore workstation equipped with several graphics processing units (GPUs). To overcome the low memory size limitation of the GPUs, and at the same time to accelerate the generation process, we propose to generate the large sparse linear systems arising in finite-element analysis in an iterative manner on several GPUs and to use the graphics...
-
GPU based implementation of Temperature-Vegetation Dryness Index for AVHRR3 Satellite Data
PublicationPaper presents an implementation of TVDI (Temperature-Vegetation-Dryness Index) algorithm on GPU (Graphics Processing Unit). Calculation of this index is based on LST (Land Surface Temperature) and NDVI (Normalized Difference Vegetation Index). Discussed results are based on multi-spectral imagery retrieved from AVHRR3 sensors for area of Poland. All phases of TVDI implementation on GPU are modified in respect to CUDA platform....
-
Multi-core and Multiprocessor Implementation of Numerical Integration in Finite Element Method
PublicationThe paper presents techniques for accelerating a numerical integration process which appears in the Finite Element Method. The acceleration is achieved by taking advantages of multi-core and multiprocessor devices. It is shown that using multi-core implementation with OpenMP and a GPU acceleration using CUDA architecture allows one to achieve the speedups by a factor of 5 and 10 on a CPU and GPUs, respectively.
-
Preconditioners with Low Memory Requirements for Higher-Order Finite-Element Method Applied to Solving Maxwell’s Equations on Multicore CPUs and GPUs
PublicationThis paper discusses two fast implementations of the conjugate gradient iterative method using a hierarchical multilevel preconditioner to solve the complex-valued, sparse systems obtained using the higher order finite-element method applied to the solution of the time-harmonic Maxwell equations. In the first implementation, denoted PCG-V, a classical V-cycle is applied and the system of equations on the lowest level is solved...
-
GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM
PublicationThis paper presents a GPU-accelerated implementation of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method with an inexact nullspace filtering approach to find eigenvalues in electromagnetics analysis with higherorder FEM. The performance of the proposed approach is verified using the Kepler (Tesla K40c) graphics accelerator, and is compared to the performance of the implementation based on functions from...
-
Tuning matrix-vector multiplication on GPU
PublicationA matrix times vector multiplication (matvec) is a cornerstone operation in iterative methods of solving large sparse systems of equations such as the conjugate gradients method (cg), the minimal residual method (minres), the generalized residual method (gmres) and exerts an influence on overall performance of those methods. An implementation of matvec is particularly demanding when one executes computations on a GPU (Graphics...
-
Towards an efficient multi-stage Riemann solver for nuclear physics simulations
PublicationRelativistic numerical hydrodynamics is an important tool in high energy nuclear science. However, such simulations are extremely demanding in terms of computing power. This paper focuses on improving the speed of solving the Riemann problem with the MUSTA-FORCE algorithm by employing the CUDA parallel programming model. We also propose a new approach to 3D finite difference algorithms, which employ a GPU that uses surface memory....
-
GPU Acceleration of Multilevel Solvers for Analysis of Microwave Components With Finite Element Method
PublicationThe letter discusses a fast implementation of the conjugate gradient iterative method with ${rm E}$-field multilevel preconditioner applied to solving real symmetric and sparse systems obtained with vector finite element method. In order to accelerate computations, a graphics processing unit (GPU) was used and significant speed-up (2.61 fold) was achieved comparing to a central processing unit (CPU) based approach. These results...
-
A Regular Expression Matching Application with Configurable Data Intensity for Testing Heterogeneous HPC Systems
PublicationModern High Performance Computing (HPC) systems are becoming increasingly heterogeneous in terms of utilized hardware, as well as software solutions. The problems, that we wish to efficiently solve using those systems have different complexity, not only considering magnitude, but also the type of complexity: computation, data or communication intensity. Developing new mechanisms for dealing with those complexities or choosing an...
-
A memory efficient and fast sparse matrix vector product on a Gpu
PublicationThis paper proposes a new sparse matrix storage format which allows an efficient implementation of a sparse matrix vector product on a Fermi Graphics Processing Unit (GPU). Unlike previous formats it has both low memory footprint and good throughput. The new format, which we call Sliced ELLR-T has been designed specifically for accelerating the iterative solution of a large sparse and complex-valued system of linear equations arising...
-
Acceleration of Electromagnetic Simulations on Reconfigurable FPGA Card
PublicationIn this contribution, the hardware acceleration of electromagnetic simulations on the reconfigurable field-programmable-gate-array (FPGA) card is presented. In the developed implementation of scientific computations, the matrix-assembly phase of the method of moments (MoM) is accelerated on the Xilinx Alveo U200 card. The computational method involves discretization of the frequency-domain mixed potential integral equation using...
-
Optymalizacja wydajności obliczeniowej metody elementów skończonych w architekturze CUDA
PublicationCelem niniejszej rozprawy oraz stypendium odbytego w ramach projektu było opracowanie numerycznie efektywnego rozwiązania algorytmicznego i sprzętowego, które umożliwia przyspieszenie analizy problemów elektromagnetycznych metodą elementów skończonych (MES) z funkcjami bazowymi wysokiego rzędu. Metoda elementów skończonych w dziedzinie częstotliwości stanowi wydajne i uniwersalne narzędzie analizy układów mikrofalowych (rys....
-
Expedited EM-Driven Design of Miniaturized Microwave Hybrid Couplers Using Surrogate-Based Optimization
PublicationMiniaturization of microwave hybrid couplers is important for contemporary wireless communication engineering. Using standard computer-aided design methods for development of compact structures is extremely challenging due to a general lack of computationally efficient and accurate simulation models. Poor accuracy of available equivalent circuits results from neglecting parasitic cross-couplings that greatly affect the performance...
-
Modelowanie wydajności, niezawodności i zużycia energii wilopoziomowych systemów równoległych wielkiej skali z uwzględnieniem CPU oraz GPU
ProjectsProject realized in Faculty of Electronics, Telecommunications and Informatics according to UMO-2012/07/B/ST6/01516 agreement from 2013-07-17
-
Food Classification from Images Using a Neural Network Based Approach with NVIDIA Volta and Pascal GPUs
PublicationIn the paper we investigate the problem of food classification from images, for the Food-101 dataset extended with 31 additional food classes from Polish cuisine. We adopted transfer learning and firstly measured training times for models such as MobileNet, MobileNetV2, ResNet50, ResNet50V2, ResNet101, ResNet101V2, InceptionV3, InceptionResNetV2, Xception, NasNetMobile and DenseNet, for systems with NVIDIA Tesla V100 (Volta) and...
-
Zastosowanie technologii GPGPU do wspomagania inżynierskich obliczeń numerycznych na przykładzie analizy przepływu przez ośrodek dwufazowy płyn - ciało stałe
PublicationW artykule po przedstawieniu podstawowych informacji na temat technologii GPGPU oraz struktury NVIDIA CUDA opisano równania zachowania rządzące przepływami oraz ich dyskretyzację numeryczna. Następnie zbadano możliwości wykorzystania technologii GPGPU w celu zoptymalizowania czasu wykonywania obliczeń numerycznych przepływu przez ośrodek dwufazowy (płyn - cząsteczki ciała stała stałego) zbliżony do ośrodka porowatego. W tym celu,...
-
Minimizing Distribution and Data Loading Overheads in Parallel Training of DNN Acoustic Models with Frequent Parameter Averaging
PublicationIn the paper we investigate the performance of parallel deep neural network training with parameter averaging for acoustic modeling in Kaldi, a popular automatic speech recognition toolkit. We describe experiments based on training a recurrent neural network with 4 layers of 800 LSTM hidden states on a 100-hour corpora of annotated Polish speech data. We propose a MPI-based modification of the training program which minimizes the...
-
Zastosowanie technologii GPGPU do wspomagania inżynierskich obliczeń numerycznych na przykładzie analizy przepływu przez ośrodek dwufazowy płyn-ciało stałe
PublicationW artykule po przedstawieniu podstawowych informacji na temat technologii GPGPU oraz struktury NVIDIA CUDA opisano równania zachowania rządzące przepływami oraz ich dyskretyzację numeryczna. Następnie zbadano możliwości wykorzystania technologii GPGPU w celu zoptymalizowania czasu wykonywania obliczeń numerycznych przepływu przez ośrodek dwufazowy (płyn - cząsteczki ciała stała stałego) zbliżony do ośrodka porowatego. W tym celu,...
-
Comparing Apples and Oranges: A Mobile User Experience Study of iOS and Android Consumer Devices
PublicationWith the rapid development of wireless networks and the spread of broadband access around the world, the number of active mobile user devices continues to grow. Each year more and more terminals are released on the market, with the smartphone being the most popular among them. They include low-end, mid-range, and of course high-end devices, with top hardware specifications. They do vary in build quality, utilized type of material,...
-
Advanced Potential Energy Surfaces for Molecular Simulation
PublicationAdvanced potential energy surfaces are defined as theoretical models that explicitly include many-body effects that transcend the standard fixed-charge, pairwise-additive paradigm typically used in molecular simulation. However, several factors relating to their software implementation have precluded their widespread use in condensed-phase simulations: the computational cost of the theoretical models, a paucity of approximate models...
-
Hybrid model of geared rotor system
PublicationIn the paper a hybrid model of a geared multirotor system has been developed. The model is obtained by application of both the modal decomposition methodology and the spatial discretization method. Reduced modal model was constructed for the system without gyroscopic and damping effects. The gyroscopic interaction, damping and other phenomena which are difficult to include in the modal approach were modeled by application of simply...
-
Hybrid Approach to Networked Control System
PublicationEffcient control of Networked Control System (NCS) is a challenge, as the control methods need to deal with non-deterministic variable delays and data loss. This paper presents a novel hybrid approach to NCS where Model Predictive Control (MPC) is applied as a main controller and implicit switching MPC is used for data transmission control in event-driven shared communication medium, leading to complex control system with active...
-
Hybrid System for Ship-Aided Design Automation
PublicationA hybrid support system for ship design based on the methodology of CBR with some artificial intelligence tools such as expert system Exsys Developer along with fuzzy logic, relational Access database and artificial neural network with backward propagation of errors.
-
Hybrid Reduced Model of Continuous System
PublicationThe paper introduces an alternative method of modelling and modal reduction of continuous systems. Presented method is a hybrid one. It combines the advantages of modal decomposition method and the rigid finite element method. In the proposed method continuous structure is divided into one-dimensional continuous elements. For each 1D element modal decomposition and reduction is applied. Interactions between substructures are...
-
Hybrid storage management system consisting of supercapacitors and AHI batteries
PublicationIn the article the hybrid storage, composed of Aqueous Hybrid Ion battery and supercapacitor was presented. The system was tested through simulation studies. The work also describes the construction and properties of the developed device meant to control the energy flow of the storage and a communication system for remote parameter supervision.
-
HYBRID ENERGY SYSTEM FOR A CLASSIC SHIP POWER PLANT
PublicationThe article presents a brief overview of hybrid energy systems used on ships. The area of their application is outlined. The benefits of using such systems are also indicated. Then, the classic ship power plant is defined. The most important part of the article is a proposal how to modify a classic engine room by using a hybrid energy system. The idea is: to accumulate a part of electricity in areas where it is allowed to burn...
-
Optimizing the parameters of a small standalone hybrid power system
PublicationA hybrid power plant consists of renewable energy resources, an energy storage, a discharge load and an emergency power supply. Power plant parameters are tailored to meet the requirements of continuity of supply, cost minimization, return on investment period and system capacity utilization. The papaer presents the methodology for selecting power plant parameters with a larger number of decision criteria. The task is solved...
-
Hybrid Expert System for Computer-Aided Design of Ship Thruster Subsystems
PublicationThe article presents an expert system supporting the design of ship's power subsystems, in particular the thruster subsystem. The proposed hybrid expert system uses the results of simulation tests as the additional source of knowledge. The results of system operation are collated in a report which can be used as part of ship design description. The work oriented on developing the expert system is the continuation of the research...
-
An interactive system for remote modeling and design validation of hybrid photovoltaic systems
PublicationIn the paper a multi-functional demonstrator of the interactive system designed to modeling, monitoring and validation of hybrid photovoltaic systems assisted by fuel cells and thermoelectric generators is presented. The purpose of this paper is to report the system solution expressed in the form of a block diagram. Technical parameters of demonstrator components such as: silicon photovoltaic modules, fuel cells, thermoelectric...
-
AN INTELLIGENT SYSTEM TO SUPPORT THE DESIGN PROCESS OF HYBRID PHOTOVOLTAIC INSTALLATIONS
PublicationThe paper presents the final version of intelligent system to support the design process of hybrid photovoltaic systems supported by fuel cells and / or thermoelectric generators developed and constructed at the Institute of Electron Technology, as well as to conduct experiments and research. A block diagram of the system and the selection of its components is described and discussed. The paper considers the availability of the...