high performance computing on graphics processing units: hgpu.org

Markovian models can generate very large sparse matrices, which are difficult to store and solve. A useful method for finding transient probabilities in Markovian models is the uniformization. The aim of this paper is to show that the performance of the uniformization can be improved using multiGPU architecture. We propose partitioning scheme for HYB sparse […]

Understanding protein-protein and protein-DNA interactions is key to understanding the dynamics of gene regulation [3,17]. We here review a previously presented method[1,15,20], based on a variation of microarray expression profile correlation analysis, that seeks to identify interactions between a putative heteropolymeric transcription factor(TF) complex and DNA as well as some experimental results that bolster the […]

Graphics processing units (GPU) have potential to speed up deep packet inspection (DPI) by processing many packets in parallel. However, popular methods of DPI such as deterministic finite automata are limited because they are single stride. Alternatively, the complexity of multiple stride methods is not appropriate for the SIMD operation of a GPU. In this […]

We implement a real-time image object tracking system with PTZ cameras. In general, mean shift algorithm is efficient for real-time tracking because of its fast and stable performance. However, in the image tracking system for PTZ cameras, the speed is not satisfied. So in this paper, we use parallel mean shift algorithm based on the […]

Many GPU parallelizations exist to speedup simulation of complex systems, but these approaches see less benefit when the simulation is not large. Simulation of many independent complex systems is useful for Monte Carlo sampling or for exploring the behavior of many different models at once. We present and evaluate an algorithm for simulating many mid-sized […]

This work revises the original file-wise reliability scheme to cope with larger pages in storage devices nowadays, and implements it as a file system prototype: CRSFS. There are four layers in CRSFS: GPU primitive for Cauchy Reed-Solomon (CRS) coding, CrystalGPU framework, CRS coding layer and AFS FUSE layer. CRSFS provides GPU acceleration on the CRS […]

Clusters using multiple GPUs have been already widespread to build a high performance computer economically. However, since the number of plugged GPUs into a CPU is limited, such clusters are consisting of multiple host PCs each of which has a few GPUs. This conventional multi-GPU cluster requires programmers to learn parallel programming skills for controlling […]

This thesis details both the technical and theoretical aspects of performing path integrals through classical Molecular Dynamics (MD) simulations. In particular, Graphics Processing Unit (GPU) computing is used to augment the Path Integral Molecular Dynamics (PIMD) portion of the widely available Molecular Modelling Tool Kit (MMTK) library. This same PIMD code is also extended in […]

Here we present a comparative insight of the performance of the Heston stochastic volatility model on different acceleration platforms. This model was tested against a MacBook’s CPU, a Techila grid server hosted on Microsoft’s Azure cloud, a GPU node hosted by Boston Ltd, and an FPGA node hosted by Maxeler Technologies Ltd. Temporal data was […]

Multicore-GPU platforms are now common and affordable, yet capitalising on their parallel processing capability is not straightforward. Existing sequential and parallel software must be tuned, or designed anew, to efficiently capitalise on these platforms. This paper presents the design of parallel data list processing in multicore-GPU platforms, wherein application data is organised into various lists, […]

We present a technique for verifying race- and divergencefreedom of GPU kernels that are written in mainstream kernel programming languages such as OpenCL and CUDA. Our approach is founded on a novel formal operational semantics for GPU programming termed synchronous, delayed visibility (SDV) semantics. The SDV semantics provides a precise definition of barrier divergence in […]

In this paper, we consider the edge detection problem using an intelligent approach. We use a multilayer neural network based on multi-valued neurons (MLMVN) as an intelligent edge enhancer. MLMVN is a complex-valued neural network and it has many advantages over classical neural networks. It significantly outperforms a classical multilayer feedforward neural network in terms […]