We give an overview of QPACE 2, which is a custom-designed supercomputer based on Intel Xeon Phi processors, developed in a collaboration of Regensburg University and Eurotech. We give some general recommendations for how to write high-performance code for the Xeon Phi and then discuss our implementation of a domain-decomposition-based solver and present a number […]

February 22, 2015 by hgpu

Obtaining a thermodynamically accurate phase diagram through numerical calculations is a computationally expensive problem that is crucially important to understanding the complex phenomena of solid state physics, such as superconductivity. In this work we show how this type of analysis can be significantly accelerated through the use of modern GPUs. We illustrate this with a […]

January 2, 2015 by hgpu

This paper describes some applications of GPU acceleration in ab initio nuclear structure calculations. Specifically, we discuss GPU acceleration of the software package MFDn, a parallel nuclear structure eigensolver. We modify the matrix construction stage to run partly on the GPU. On the Titan supercomputer at the Oak Ridge Leadership Computing Facility, this produces a […]

December 22, 2014 by hgpu

We adopt CUDA-capable Graphic Processing Units (GPUs) for Landau, Coulomb and maximally Abelian gauge fixing in 3+1 dimensional SU(3) and SU(2) lattice gauge field theories. A combination of simulated annealing and overrelaxation is used to aim for the global maximum of the gauge functional. We use a fine grained degree of parallelism to achieve the […]

December 12, 2014 by hgpu

We describe a highly optimized implementation of MPI domain decomposition in a GPU-enabled, general-purpose molecular dynamics code, HOOMD-blue (Anderson and Glotzer, arXiv:1308.5587). Our approach is inspired by a traditional CPU-based code, LAMMPS (Plimpton, J. Comp. Phys. 117, 1995), but is implemented within a code that was designed for execution on GPUs from the start (Anderson […]

December 12, 2014 by hgpu

The gap between the cost of moving data and the cost of computing continues to grow, making it ever harder to design iterative solvers on extreme-scale architectures. This problem can be alleviated by alternative algorithms that reduce the amount of data movement. We investigate this in the context of Lattice Quantum Chromodynamics and implement such […]

December 9, 2014 by hgpu

In this work, we present the GPU implementation of the overrelaxation and steepest descent method with Fourier acceleration methods for Laudau and Coulomb gauge fixing using CUDA for SU(N) with N>2. A multi-GPU implementation of the overrelaxation method is also presented using MPI and CUDA. The GPU performance was measured on BlueWaters and compared against […]

December 5, 2014 by hgpu

Electrical power requirements will be a constraint on the future growth of Distributed High Throughput Computing (DHTC) as used by High Energy Physics. Performance-per-watt is a critical metric for the evaluation of computer architectures for cost- efficient computing. Additionally, future performance growth will come from heterogeneous, many-core, and high computing density platforms with specialized processors. […]

December 5, 2014 by hgpu

As a generic example for crystals where the crystal-fluid interface tension depends on the orientation of the interface relative to the crystal lattice axes, the nearest neighbor Ising model on the simple cubic lattice is studied over a wide temperature range, both above and below the roughening transition temperature. Using a thin film geometry $L_x […]

November 25, 2014 by hgpu

Lattice Quantum Chromodynamics simulations typically spend most of the runtime in inversions of the Fermion Matrix. This part is therefore frequently optimized for various HPC architectures. Here we compare the performance of the Intel Xeon Phi to current Kepler-based NVIDIA Tesla GPUs running a conjugate gradient solver. By exposing more parallelism to the accelerator through […]

November 18, 2014 by hgpu

A micromagnetic simulator running on graphics processing unit (GPU) is presented. It achieves significant performance boost as compared to previous central processing unit (CPU) simulators, up to two orders of magnitude for large input problems. Different from GPU implementations of other research groups, this simulator is developed with C++ Accelerated Massive Parallelism (C++ AMP) and […]

November 12, 2014 by hgpu

Much work has recently been reported in parallel GPU-based particle swarm optimization (PSO). Motivated by the encouraging results of these investigations, while also recognizing the limitations of GPU-based methods for big problems using a large amount of data, this paper explores the efficacy of employing other types of parallel hardware for PSO. Most commodity systems […]

October 20, 2014 by hgpu