high performance computing on graphics processing units: hgpu.org

A leading-order, leading-color parton-level event generator is developed for use on a multi-threaded GPU. Speed-up factors between 150 and 300 are obtained compared to an unoptimized CPU-based implementation of the event generator. In this first paper we study the feasibility of a GPU-based event generator with an emphasis on the constraints imposed by the hardware. […]

A graphic processing unit (GPU) implementation of the multicomponent lattice Boltzmann equation with multirange interactions for soft-glassy materials [“glassy” lattice Boltzmann (LB)] is presented. Performance measurements for flows under shear indicate a GPU/CPU speed up in excess of 10 for 1024 2 grids. Such significant speed up permits to carry out multimillion time-steps simulations of […]

Deep shadow maps unify the computation of volumetric and geometric shadows. For each pixel in the shadow map, a fractional visibility function is sampled, pre-filtered, and compressed as a piecewise linear function. However, the original implementation targets software-based off-line rendering. Similar previous algorithms on GPUs focus on geometric shadows and lose many important benefits of […]

Despite the recent advances in graphics hardware capabilities, a brute force approach is incapable of interactively displaying terabytes of data. We have implemented a system that uses hierarchical level-of-detailing for the results of cosmological simulations, in order to display visually accurate results without loading in the full dataset (containing over 10 billion points). The guiding […]

In this paper, we introduce an innovative CNN algorithm development environment that significantly assists algorithmic design. The introduced graphical user interface uses Matlab Simulink with UMF-like program description, where direct functionality accompanies better accessability. The new generation of graphical cards incorporate many general purpose graphics processing units, giving the power of parallel computing to a […]

In this paper we describe a GPU parallelization of the 3D finite difference computation using CUDA. Data access redundancy is used as the metric to determine the optimal implementation for both the stencil-only computation, as well as the discretization of the wave equation, which is currently of great interest in seismic computing. For the larger […]

Sorting is a well-investigated topic in Computer Science in general and by now many efficient sorting algorithms for CPUs and GPUs have been developed. There is no swapping, paging, etc. available on GPUs to provide more virtual memory than physically available, thus if one wants to sort sequences that exceed GPU memory using the GPU […]

In this work, the emphasis is on the development of strategies to realize techniques of numerical computing on the graphics chip. In particular, the focus is on the acceleration of techniques for solving sets of algebraic equations as they occur in numerical simulation. We introduce a framework for the implementation of linear algebra operators on […]

In this paper we discuss about our experiences in improving the performance of two key algorithms: the single-precision matrix-matrix multiplication subprogram (SGEMM of BLAS) and single-precision FFT using CUDA. The former is computation-intensive, while the latter is memory bandwidth or communication-intensive. A peak performance of 393 Gflops is achieved on NVIDIA GeForce GTX280 for the […]

The Finite-Difference Time-Domain (FDTD) method is used extensively in areas of microwave engineering and optics. However, FDTD runs too slow for some simulations to be practical, especially when run on standard desktop computers. The suitability of dedicated hardware for the acceleration of FDTD computations has been investigated. It is demonstrated that standard consumer Graphics Processor […]

Control statements in a GPU program such as loops and branches pose serious challenges for the efficient usage of GPU resources because those control statements will lead to the serialization of threads and consequently ruin the occupancy of GPU, that is, the number of threads running concurrently. Unlike traditional vector processing units that are inside […]

Although pixel shaders were designed for the creation of programmable rendering effects, they can also be used as generic processing units for vector data. In this paper, attention is paid to an implementation of the YCoCg-R to RGB color space transform, as defined in the H.264/AVC Fidelity Range Extensions, by making use of pixel shaders. […]