GPGPU Workshop at ICCS 2006

Accepted papers

This is a list of the accepted papers as they are published in the LNCS proceedings.

GPU Accelerated Smith-Waterman

• Authors: Y. Lui, W. Huang, J. Johnson, S. Vaidya
• Affiliation: Lawrence Livermore National Laboratory, United States
• Abstract: We present a novel hardware implementation of the double-affine Smith-Waterman (DASW) algorithm, which uses dynamic programming (DP) to compare and align genomic sequences such as DNA and proteins. We implement DASW on a commodity graphics card, taking advantage of the general purpose programmability of the graphics processing unit (GPU) to leverage its cheap parallel processing power. The results demonstrate that the performance of our system is competitive with current optimized software packages.
• Citation: bibtex

A Graphics Hardware Accelerated Algorithm for Nearest Neighbor Search

• Authors: B. Bustos, O. Deussen, S. Hiller, D. Keim
• Affiliation: University of Konstanz, Germany
• Abstract: We present a GPU algorithm for the nearest neighbor search, an important database problem. The search is completely performed using the GPU: No further post-processing using the CPU is needed. Our experimental results, using large synthetic and real-world data sets, showed that our GPU algorithm is several times faster than its CPU version.
• Citation: bibtex

The Development of the Data-Parallel GPU Programming Language CGiS

• Authors: P. Lucas, N. Fritz, R. Wilhelm
• Affiliation: Universitaet des Saarlandes
• Website: http://rw4.cs.uni-sb.de/~phlucas/pubs/GPGPU06.html
• Abstract: In this paper, we present the recent developments on the design and implementation of the data-parallel programming language CGiS. CGiS is devised to facilitate use of the data-parallel resources of current graphics processing units (GPUs) for scientific programming.
• Citation: bibtex

Spline Surface Intersections Optimized for GPUs

• Authors: S. Briseid, T, Dokken, T.R. Hagen, J.O. Nygaard
• Affiliation: SINTEF Applied Mathematics, Norway
• Website: http://www.sintef.no/gpgpu
• Abstract: A commodity-type graphics card with its graphics processing unit (GPU) is used to detect, compute and visualize the intersection of two spline surfaces, or the self-intersection of a single spline surface. The parallelism of the GPU facilitates fast and efficient subdivision and bounding box testing of smaller spline patches and their corresponding normal subpatches. This subdivision and testing is iterated until a prescribed level of accuracy is reached, after which results are returned to the main computer. We observe speedups up to $17$ times relative to a contemporary 64 bit CPU.
• Citation: bibtex

A GPU Implementation of Level Set Multiview Stereo

• Authors: P. Labatut, R. Keriven, J.-P. Pons
• Affiliation: Département d'Informatique, École normale supérieure; and CERTIS, École Nationale des Ponts et Chaussées; France
• Abstract: Variational methods that evolve surfaces according to PDEs have been quite successful for solving the multiview stereo shape reconstruction problem since [1]. However just like every other algorithm that tackles this problem, their running time is quite high (from dozen of minutes to several hours). Fortunately graphics hardware has shown a great potential for speeding up many low-level computer vision tasks. In this paper, we present the analysis of the different bottlenecks of the original implementation of [2] and show how to efficently port it to GPUs using well-known GPGPU techniques. We finally present some results and discuss the improvements.
• Citation: bibtex

Solving the Euler Equations on Graphics Processing Units

• Authors: T.R. Hagen, K.-A. Lie, J.R. Natvig
• Affiliation: SINTEF, Norway
• Website: http://www.sintef.no/gpgpu
• Abstract: The paper describes how one can use commodity graphics cards (GPUs) as a high-performance parallel computer to simulate the dynamics of ideal gases in two and three spatial dimensions. The dynamics is described by the Euler equations, and numerical approximations are computed using state-of-the-art high-resolution finite-volume schemes. These schemes are based upon an explicit time discretisation and are therefore ideal candidates for parallel implementation.
• Citation: bibtex

Particle-Based Fluid Simulation on the GPU

• Authors: K. Hegeman, N. Carr, G. Miller
• Affiliation: Stony Brook University, Adobe Systems Inc., United States
• Abstract: Large scale particle-based fluid simulation is important to both the scientific and computer graphics communities. In this paper, we explore the effectiveness of implementing smoothed particle hydrodynamics on the streaming architecture of a GPU. A dynamic quadtree structure is proposed to accelerate the computation of inter-particle forces. Our method readily extends to higher dimensions without undue increase in memory or computation costs. We show that a GPU implementation runs nearly an order of magnitude faster than our CPU version for large problem sizes.
• Citation: bibtex

Spiking Neurons on GPUs

• Authors: F. Bernhard, R. Keriven
• Affiliation: Projet Odyssée - ENS, Projet Odyssée- ENS - ENPC, France
• Abstract: Simulating large networks of spiking neurons is a very common task in the areas of Neuroinformatics and Computational Neurosciences. These simulations are time-consuming but also often intrinsically parallel. The recent advent of powerful and programmable graphic cards seems to be a pertinent solution to the problem: they offer a cheap but efficient possibility to serve as very fast co-processors for the parallel computing that spiking neural networks need. We describe our implementation of three different problems on such a card: two image-segmentation algorithms using spiking neural networks and one multi-purpose spiking neural-network simulator. Using these examples we show the benefits, the challenges and the limits of such an implementation.
• Citation: bibtex