high performance computing on graphics processing units: hgpu.org

In this work, we present the concept, design and implementation of a new software to visualize and segment 3-dimensional medical data. The main goal was to create a platform that would allow to try out new approaches and ideas while staying independent from hardware and operating system, being especially useful for interdisciplinary research groups. A […]

Before an intervention on a nuclear site, it is essential to study different scenarios to identify the less dangerous one for the operator. Therefore, it is mandatory to dispose of an efficient dosimetry simulation code with accurate results. One classical method in radiation protection is the straight-line attenuation method with build-up factors. In the case […]

We discuss finite volume errors in our calculations of $B_K$ using improved staggered fermions on the MILC asqtad lattices. Using GPUs, we are now able to extrapolate using next-to-leading order (NLO) staggered SU(2) chiral perturbation theory including the finite volume corrections arising from pion loops. We find that the impact of FV fitting is very […]

GPU has a significantly higher performance in single-precision computing than that of double precision. Hence, it is important to take a maximal advantage of the single precision in the CG inverter, using the mixed precision method. We have implemented mixed precision algorithm to our multi GPU conjugate gradient solver. The single precision calculation use half […]

In this paper we show how it is possible to improve the efficiency of existing holistic forms of object detection by refining detection areas to smaller subsets. Although this method can be applied to any form of object detection, this paper will specifically focus on the topic of pedestrian detection in lowresolution non-stationary video footage.

With the ubiquity of multicore computing, and the likely expansion of it, it seems irresponsible for constraints researchers to ignore the implications of it. Therefore, the authors have recently begun investigating the literature in constraints on exploitation of parallel systems for constraint solving. We have been compiling an incomplete, biased, and ill-written review of this […]

Using multi-GPU systems, including GPU clusters, is gaining popularity in scientific computing. However, when using multiple GPUs concurrently, the conventional data parallel GPU programming paradigms, e.g., CUDA, cannot satisfactorily address certain issues, such as load balancing, GPU resource utilization, overlapping fine grained computation with communication, etc. In this paper, we present a fine-grained task-based execution […]

In this paper, an energy efficient architecture for Build Energy Efficient GPU and CPU Cluster Using DRBL is proposed. This architecture helps administrator not only to quickly deploy and manage GPU and CPU Cluster environment, but also bring benefit of energy efficiency in scientific computing. The experiment simulates 3 cases to prove energy efficiency. We […]

he iterative solution of the coupled-cluster with single and double excitations (CCSD) equations is a very time-consuming component of the "gold standard" in quantum chemistry, the CCSD(T) method. In an effort to accelerate accurate quantum mechanical calculations, we explore two implementation strategies for the iterative solution of the CC equations on graphics procesing units (GPUs). […]

Differential evolution is an efficient meta-heuristic optimization method with solid record of real world applications. In this paper, we present a simple and efficient implementation of the differential evolution using the massively parallel CUDA architecture. We demonstrate the speedup and improvements obtained by the parallelization of this intelligent algorithm on the problem of scheduling of […]

With a variety of processing technologies available today, using a combination of different technologies often provides the best performance for a particular task. However, unifying multiple processors with different instruction sets can be a very ad hoc and difficult process. The Open Component Portability Infrastructure (OpenCPI) provides a platform that simplifies programming heterogeneous processing applications […]

Scattering maps from strained or disordered nanostructures around a Bragg reflection can be either computed quickly using approximations and a (fast) Fourier transform or obtained using individual atomic positions. In this article, it is shown that it is possible to compute up to 4*10^10 reflections*atoms*s^-1 using a single graphics card, and the manner in which […]