high performance computing on graphics processing units: hgpu.org

Cone beam CT (CBCT) has been widely used for patient setup in image guided radiation therapy (IGRT). Radiation dose from CBCT scans has become a clinical concern. The purposes of this study are 1) to commission a GPU-based Monte Carlo (MC) dose calculation package gCTD for Varian On-Board Imaging (OBI) system and test the calculation […]

In various versions of WINRAR, the file security is mainly protected by user authentication and files encryption. Password based key derivation function (PBKDF) is the core of the WINRAR security mechanism. In this paper, the security of PBKDF algorithm and the encrypted file in WINRAR are analyzed by the Game-Playing approach. We show the upper […]

Feature detection and extraction are essential in computer vision applications such as image matching and object recognition. The Scale-Invariant Feature Transform (SIFT) algorithm is one of the most robust approaches to detect and extract distinctive invariant features from images. However, high computational complexity makes it difficult to apply the SIFT algorithm to mobile applications. Recent […]

This work will present the utilization of the massively multi-threaded environment of graphics processors (GPUs) to improve the computation time needed to reconstruct large computed tomography (CT) datasets and the arising challenges for system implementation. Intelligent algorithm design for massively multi-threaded graphics processors differs greatly from traditional CPU algorithm design. Although a brute force port […]

The Split-Field Finite-Difference Time-Domain (SF-FDTD) scheme is an optimal formulation for modeling periodic optical media by means of a single unit period. The split-field components and the Periodic Boundary Condition (BPC) in the periodic boundaries allow to obtain successful results even with oblique angle of incidence. Under this situation the standard FDTD scheme requires multiple […]

In this paper we revisit stencil methods on GPUs in the context of exponential integrators. We further discuss boundary conditions, in the same context, and show that simple boundary conditions (for example, homogeneous Dirichlet or homogeneous Neumann boundary conditions) do not affect the performance if implemented directly into the CUDA kernel. In addition, we show […]

This paper explores the early implementation of high-performance routines for the solution of multiple large Hermitian eigenvector and eigenvalue systems on a Graphics Processing Unit (GPU). We report a performance increase of up to two orders of magnitude over the original EISPACK routines with a NVIDIA Tesla C2050 GPU, potentially allowing an order of magnitude […]

The purpose of this thesis was to benchmark and compare different representations of sparse matrices and algorithms for multiplying them with a vector. Also, to see the performance differences of running the algorithms on a CPU and GPU(s). Four different storage formats were tested – full matrix storage, coordinate storage (COO), ELLPACK (ELL), compressed sparse […]

Software that calculate different scenarios of field development play important role in petroleum industry. Increasing number of cells in the simulation grid significantly slows down the calculations. In order to obtain accuracy results it is necessary to spend a lot of time for the simulations (days or weeks) or use expensive high-performance systems or supercomputers. […]

This study reports the GPU parallelization of complex three-dimensional software for nonlinear analysis of concrete structures. It focuses on coupled thermo-mechanical analysis of complex structures. A coupled FEM/DEM approach (CDEM) is given from a fundamental theoretical viewpoint. As the modeling of a large structure by means of FEM/DEM may lead to prohibitive computation times, a […]

The Parallel Ocean Program (POP) is used in many strongly eddying ocean circulation simulations. Ideally one would like to do thousand-year long simulations, but the current performance of POP prohibits this type of simulations. In this work, using a new distributed computing approach, two innovations to improve the performance of POP are presented. The first […]

This work presents an efficient method to map Motion Estimation (ME) algorithms onto General Purpose Graphic Processing Unit (GPGPU) architectures using CUDA programming model. Our method jointly exploits the massive parallelism available in current GPGPU devices and the parallelization potential of ME algorithms: Full Search (FS) and Diamond Search (DS). Our main goal is to […]