Tag: hybrid computing
Accelerating VASP electronic structure calculations using graphic processing units
We have provided a hybrid massively parallelized molecular dynamic VASP ab initio software for GPUs clusters. To avoid continuously transferring data from CPUs to GPUs, we have ported some functions in CUDA and achieved a balanced combination between CUFFT, CUBLAS, and CUDA.
A fully parallel, high precision, N-body code running on hybrid computing platforms
We present a new implementation of the numerical integration of the classical, gravitational, N-body problem based on a high order Hermite’s integration scheme with block time steps, with a direct evaluation of the particle-particle forces
Hybrid computing: CPU+GPU co-processing and its application to tomographic reconstruction
This work presents a hybrid approach that collaboratively combines the GPUs and CPUs available in a computer and applies it to the problem of tomographic reconstruction
GPU Systems extended New Libra SDK to Major Operating Systems and Programming Languages
The GPU Systems just released new version of the Libra Platform & SDK with enhanced support for developers to uniformly accelerate software applications across major operating systems, Windows, Mac and Linux.
Seamlessly portable applications: Managing the diversity of modern heterogeneous systems
In this work, we propose a technique to effectively decouple applications from their accelerator-specific parts, respectively code. These parts are only linked on demand and thereby an application can be made portable across systems with different accelerators.
Computer Scientists at ORNL Collect Computing Tools for Next-Generation Machines
Researchers using the OLCF’s resources can foresee substantial changes in their scientific application code development in the near future. Tools developers attempt to make change to hybrid architectures a smooth transition.
PRACE Winter School 2012: Hybrid Programming on Massively-Parallel Architectures
The school will be focused on hybrid programming for the best exploitation of massively parallel architectures. The facility available at CINECA for exercises is called PLX and is the largest public GPU cluster in Europe.
High performance computing using MPI and OpenMP on multi-core parallel systems
In this paper, we focus on a hybrid approach to programming multi-core based HPC systems, combining standardized programming models – MPI for distributed memory systems and OpenMP for shared memory systems.
Heterogeneous Computing: A New Paradigm for the Exascale Era
In a freshly published report, industry analyst firm IDC argues that the fastest path to this exaflop milestone is through heterogeneous designs. They state that x86 processors will not be enough to meet the performance and power goals that the U.S. Department of Energy (DOE) has outlined
Advanced School of Parallel Computing: Video of talks is available
The Advanced School of Parallel Computing is an intense, 5 day, graduate level course in high performance computing including CUDA, OpenCL and GPGPU programming
