This paper studies the impact of varying the population’s size and the problem’s dimensionality in a parallel implementation, for an NVIDIA GPU, of a canonical genetic algorithm.
This paper presents a parallel implementation of piecewise autoregressive modeling image interpolation algorithm, using CUDA (Compute Unified Device Architecture) on GPU
Science magazine published interesting news focus on exascale computing. Article describes challenges on the way the modern supercomputers are built and run
We show various usage scenarios of our PDL and demonstrate the effectiveness of our framework for a commonly used scientific kernel and a financial application on different configurations of a state-of-the-art CPU/GPU system.
This paper presents a number of algorithms to run the fast multipole method (FMM) on NVIDIA CUDA-capable graphical processing units. The FMM is a class of methods to compute pairwise interactions between N particles for a given error tolerance and with computational cost.
In this paper, we evaluated various aspects of using OpenCL as a performance-portable method for GPGPU application development. Profiling results showed that environment setup overhead is large and should be minimized.
In this paper, we deal with a GPU implementation of Ant Colony Optimisation (ACO), a population-based optimisation method which comprises two major stages: Tour construction and Pheromone update.
In this paper, we have described how a such a parameter space survey could be achieved using a modern GPU-based supercomputing facility, and our experiences with the GPU-D code.
In the twilight of Moore’s Law, the transitions to multicore processors, GPU computing, and HaaS cloud computing are not separate trends, but aspects of a single trend – mainstream computers from desktops to ‘smartphones’ are being permanently transformed into heterogeneous supercomputer clusters.
BEAGLE is an API and library for high-performance evaluation of phylogenetic likelihoods. The API provides a uniform interface for performing calculations on an expanding variety of computer hardware platforms including GPUs, multicore CPUs, and SSE vectorization.
Erlang is ideal for writing robust highly concurrent applications which take full advantage of today’s multi-core computers. NVIDIA’s CUDA API is great for using GPUs to rip through large datasets at incredible speeds. What happens when you put these two together?
In the video Gary Davis of Autodesk covers his 90-minute lecture from Autodesk University 2011, focusing on GPU-accelerated rendering within Autodesk 3ds Max 2012.
Featuring Steve Scott, CTO, Tesla, NVIDIA and Michael Jong, Senior Marketing Director, NVIDIA
NVIDIA today released a new version of its CUDA parallel computing platform, which will make it easier for computational biologists, chemists, physicists, geophysicists, other researchers, and engineers to advance their simulations and computational work by using GPUs.
Nvidia’s next-generation 28nm GPU architecture, codenamed Kepler, is officially expected to launch in early Q2 2012 according to the latest schedule from the company.
ICube is a fabless semiconductor company developing semiconductor System-On-Chip (SOC) solutions based on our Harmony Unified Processor Technology, genuinely integrating two different processor types: a central processing unit (CPU) and a graphics processing unit (GPU), into one unified core.
The goal of this track is to explore the use of emerging parallel computing architectures as well as High Performance Computing systems (Supercomputers, Clusters, Grids) for the simulation of relevant biological systems.
The PAPP workshop focuses on practical aspects of high-level parallel programming: design, implementation and optimisation of high-level programming languages, semantics of parallel languages, formal verification, design or certification of libraries, middle-wares and tools
aims to bring together annually researchers and scientists from mathematics and computer science as basic computing disciplines, researchers from various application areas who are pioneering advanced application of computational methods to sciences such as physics, chemistry, life sciences, and engineering
Science magazine published interesting news focus on exascale computing. Article describes challenges on the way the modern supercomputers are built and run
In the twilight of Moore’s Law, the transitions to multicore processors, GPU computing, and HaaS cloud computing are not separate trends, but aspects of a single trend – mainstream computers from desktops to ‘smartphones’ are being permanently transformed into heterogeneous supercomputer clusters.
In this paper we present an alternative approach: a new computational framework for the development of massively data parallel scientific codes applications suitable for use on such petascale/exascale hybrid systems built upon the highly scalable Cactus framework.
This paper studies the impact of varying the population’s size and the problem’s dimensionality in a parallel implementation, for an NVIDIA GPU, of a canonical genetic algorithm.
This paper presents a parallel implementation of piecewise autoregressive modeling image interpolation algorithm, using CUDA (Compute Unified Device Architecture) on GPU
Science magazine published interesting news focus on exascale computing. Article describes challenges on the way the modern supercomputers are built and run
A new OpenCL-based multi-threaded GPU and APU stressing module for the AIDA64 System Stability Test, using GPGPU computational tasks. Support for AMD CrossFireX, AMD DualGraphics and nVIDIA SLI configurations with simultaneous stressing of all available GPU and APU devices.
BEAGLE is an API and library for high-performance evaluation of phylogenetic likelihoods. The API provides a uniform interface for performing calculations on an expanding variety of computer hardware platforms including GPUs, multicore CPUs, and SSE vectorization.
The implementation presented uses OpenCL – an open-source library designed to run on any commercially available GPU and on any operating system.
What makes debugging of multiprocess and multithread applications so difficult? The first thing that comes to mind of every concurrent programmer is the lack of program execution reproducibility. The reason for such program behavior is the preemptive scheduling employed by real-time operating systems.
The paper presents the Xcelerit SDK, a parallel programming toolkit that leverages the dataflow programming model to efficiently use multi-core CPUs, graphics processors (GPUs), and combinations of these in a cluster (or grid) from a single high-level source code.
This is “programmer’s introduction” where we cover the ideas behind OpenCL but also show how these ideas are translated into source code. We will do this through a series of progressively more challenging examples
We describe how to perform preprocessing and statistical analysis of fMRI data on the GPU. Non-parametric tests of fMRI data become practically feasible by using the GPU. GPUs are required to handle the future increase in spatial and temporal resolution. GPUs enable more advanced real-time analysis.
GPU Pro3, the third volume in the GPU Pro book series, offers practical tips and techniques for creating real-time graphics that are useful to beginners and seasoned game and graphics programmers alike.
Attendees of this HP2C training event will learn about the Cray XK6 hybrid multi-core and GPU architecture and its programming environment.
On January the 25th 2012, a Meetup of the HPC GPU supercomputing group is held in the Ecole des Mines de Paris. Par4All and HPC project will be present with a talk by Ronan Keryell on automatic paralleization of C and Fortran for multicores and GPU.
In this paper we present an implementation of SS-HOPM targeted for a GPU. We describe how to exploit symmetry to save both storage and computation in the two main computational kernels of the algorithm, and for the case of solving many small tensor eigenproblems we show how to map the computation onto a GPU.
Enter your email address below to receive updates each time we publish new content.
