high performance computing on graphics processing units: hgpu.org

To aid application characterization and architecture design space exploration, researchers and engineers have developed a wide range of tools for CPUs, including simulators, profilers, and binary instrumentation tools. With the advent of GPU computing, GPU manufacturers have developed similar tools leveraging hardware profiling and debugging hooks. To date, these tools are largely limited by the […]

In this paper, we propose a new very simple numerical method for solving liquid-gas compressible flows on two dimensional cartesian meshes. For achieving high performance, the scheme is tested on recent multi-core processors and Graphics Processing Units (GPU), using the OpenCL environment. We describe how to install and to run the code CLBUBBLE for computing […]

Submission Deadline: 2015-07-10 Topics: Communications, Information and Network Security Access control Anti-malware Anonymity Applied cryptography Authentication and authorization Biometric security Data and system integrity Database security Distributed systems security Electronic commerce Fraud control Grid security Information hiding and watermarking Intellectual property protection Intrusion detection Key management and key recovery Language-based security Operating system security Network […]

Submission Deadline: 2015-07-10 Topics: Image acquisition Detection and Estimation of Signal Parameters Image processing Signal Identification Medical image processing Nonlinear Signals and Systems Pattern recognition and analysis Time-Frequency Signal Analysis Visualization Signal Reconstruction Image coding and compression Spectral Analysis Face Recognition Filter Design and Structures Super-resolution imaging FIR Filters Image segmentation IIR Filters Face recognition […]

Submission Deadline: 2015-07-10 Topics: • Evolutionary Robotics • Distributed Sensor Networks • Robot Surgery • Search and Rescue Robots • Biorobotics • Humanoid Robotics • Autonomous Vehicles • Entertainment Robots • Rehabilitation Robotics • Micro/Nano Robotics • Underwater Robots • Service Robotics • Sensors and Early Vision • Color and Texture • Segmentation and Grouping […]

General sparse matrix-matrix multiplication (SpGEMM) is a fundamental building block for numerous applications such as algebraic multigrid method (AMG), breadth first search and shortest path problem. Compared to other sparse BLAS routines, an efficient parallel SpGEMM implementation has to handle extra irregularity from three aspects: (1) the number of nonzero entries in the resulting sparse […]

This paper presents a multi-swarm PSO algorithm for the Quadratic Assignment Problem (QAP) implemented on OpenCL platform. Our work was motivated by results of time efficiency tests performed for single-swarm algorithm implementation that showed clearly that the benefits of a parallel execution platform can be fully exploited, if the processed population is large. The described […]

Having learned a great deal about the problem and also the solutions over the course of this project, it is the opinion of the author that the method undertaken within this report is unsatisfactory for delivering performance enhancement over alternative approaches. Firstly the domain transfers result in reduced performance. For larger simulations these prove to […]

We present a multi-GPU graph processing library that allows programmers to easily extend single-GPU graph algorithms to achieve scalable performance on large graph datasets with billions of edges. Our design only requires users to specify a few algorithm-dependent blocks, hiding most multi-GPU related implementation details. Our design effectively overlaps computation and data transfer and implements […]

Deep convolutional neural networks (CNN) have recently been shown in many computer vision and pattern recognition applications to outperform by a significant margin state-of-the-art solutions that use traditional hand-crafted features. However, this impressive performance is yet to be fully exploited in robotics. In this paper, we focus one specific problem that can benefit from the […]

Recent trends of aggressive technology scaling have greatly exacerbated the occurrences and impact of faults in computing systems. This has made `reliability’ a first-order design constraint. To address the challenges of reliability, several techniques have been proposed. This paper provides a survey of architectural techniques for improving resilience of computing systems. We especially focus on […]

We present Caffe con Troll (CcT), a fully compatible end-to-end version of the popular framework Caffe with rebuilt internals. We built CcT to examine the performance characteristics of training and deploying general-purpose convolutional neural networks across different hardware architectures. We find that, by employing standard batching optimizations for CPU training, we achieve up to one […]