本文介绍了一种用于振动分子光谱计算的半经典初值表示(SC-IVR)传播子的图形处理单元(GPU)实现。时间平均制定的SC-IVR功率谱计算。提供了关于半经典代码的GPU实现的细节。考虑了原子数量不断增加的四种分子,GPU计算的振动频率与基准值完全匹配。讨论了两个GPU(NVIDIA Tesla C2075和Kepler K20)分别与两个CPU(Intel Core i5和Intel Xeon E5- 2687 W)的计算时间缩放以及与GPU实现相关的关键问题。计算时间和功耗的减少是显着的,半经典GPU计算被证明是环境友好的。
英文摘要:
This paper presents a Graphics Processing Units (GPUs) implementation of the Semiclassical Initial Value Representation (SC-IVR) propagator for vibrational molecular spectroscopy calculations. The time-averaging formulation of the SC-IVR for power spectrum calculations is employed. Details about the GPU implementation of the semiclassical code are provided. Four molecules with an increasing number of atoms are considered and the GPU-calculated vibrational frequencies perfectly match the benchmark values. The computational time scaling of two GPUs (NVIDIA Tesla C2075 and Kepler K20) respectively versus two CPUs (Intel Core i5 and Intel Xeon E5-2687W) and the critical issues related to the GPU implementation are discussed. The resulting reduction in computational time and power consumption is significant and semiclassical GPU calculations are shown to be environment friendly.
Accelerated Event-by-Event Neutrino Oscillation Reweighting with Matter Effects on a GPU
在现代中微子振荡分析中,振荡几率的计算越来越占用CPU。重新加权的Monte Carlo样本中的各个事件的独立性使其自身适合于在图形处理单元上的并行实现。使用CUDA C API将库“Prob3++”移植到GPU,允许通过恒定密度的物质对中微子振荡概率进行大规模并行计算,与单个CPU上的性能相比,将执行时间减少了75倍。
We present the Tucker tensor DFT (TTDFT) code which uses a tensor-structured algorithm with graphic processing unit (GPU) acceleration for conducting ground-state DFT calculations on large-scale systems. The Tucker tensor DFT algorithm uses a localized Tucker tensor basis computed from an additive separable approximation to the Kohn-Sham Hamiltonian. The discrete Kohn-Sham problem is solved using Chebyshev filtering subspace iteration method that relies on matrix-matrix multiplications of a sparse symmetric Hamiltonian matrix and a dense wavefunction matrix, expressed in the localized Tucker t…
This paper presents a generalised and efficient wall boundary treatment in the smoothed particle hydrodynamics (SPH) method for 3-D complex and arbitrary geometries with single- and multi-phase flows to be executed on graphics processing units (GPUs). Using a force balance between the wall and fluid particles with a novel penalty method, a pressure boundary condition is applied on the wall dummy particles which effectively prevents non-physical particle penetration into the wall boundaries also in highly violent impacts and multi-phase flows with high density ratios. A new density reinitialisa…
The Monte Carlo method is a powerful technique for computing thermodynamic magnetic states of otherwise unsolvable spin Hamiltonians, but the method becomes computationally prohibitive with increasing number of spins and the simulation of real materials and nanostructures is cumbersome. This paper presents the acceleration of Monte Carlo simulations of the three-dimensional anisotropic Heisenberg model on Graphics-Processing Units (GPU). The GPU implementation of the method presented here provides an acceleration of two orders of magnitude over conventional implementations and enables the simu…
A discontinuous Galerkin method for the discretization of the compressible Euler equations, the governing equations of inviscid fluid dynamics, on Cartesian meshes is developed for use of Graphical Processing Units via OCCA, a unified approach to performance portability on multi-threaded hardware architectures. A 30x time-to-solution speedup over CPU-only implementations using non-CUDA-Aware MPI communications is demonstrated up to 1,536 NVIDIA V100 GPUs and parallel strong scalability is shown up to 6,144 NVIDIA V100 GPUs for a problem containing 345 billion unknowns. A comparison of CUDA-Awa…
Since numerical computing with MATLAB offers a wide variety of advantages, such as easier developing and debugging of computational codes rather than lower-level languages, the popularity of this tool is significantly increased in the past decade. However, MATLAB is slower than other languages. Moreover, utilizing MATLAB parallel computing toolbox on the Graphics Processing Unit (GPU) face some limitations. The lack of attention to these limitations reduces the program execution speed. Even sometimes, parallel GPU codes are slower than serial. In this paper, some techniques in using MATLAB par…
As a first approximation beyond linearity, the nonlinear Schrödinger equation (NLSE) reliably describes a broad class of physical systems. Though numerical solutions of this model are well-established, these methods can be computationally complex. In this paper, we showcase a code development approach, demonstrating how computational time can be significantly reduced with readily available graphics processing unit (GPU) hardware and a straightforward code migration using open-source libraries. This process shows how CPU computations with power-law scaling in computation time with grid size can…