PySAGES: flexible, advanced sampling methods accelerated with GPUs
Pablo F. Zubieta Rico, Ludwig Schneider, Gustavo R. Pérez-Lemus, Riccardo Alessandri, Siva Dasetty, Cintia A. Menéndez, Yiheng Wu, Yezhi Jin, Yinan Xu, Trung D.… 展开作者
Pablo F. Zubieta Rico, Ludwig Schneider, Gustavo R. Pérez-Lemus, Riccardo Alessandri, Siva Dasetty, Cintia A. Menéndez, Yiheng Wu, Yezhi Jin, Yinan Xu, Trung D. Nguyen, John A. Parker, Andrew L. Ferguson, Jonathan K. Whitmer, Juan J. de Pablo
Molecular simulations are an important tool for research in physics, chemistry, and biology. The capabilities of simulations can be greatly expanded by providing access to advanced sampling methods and techniques that permit calculation of the relevant underlying free energy landscapes. In this sense, software that can be seamlessly adapted to a broad range of complex systems is essential. Building on past efforts to provide open-source community supported software for advanced sampling, we introduce PySAGES, a Python implementation of the Software Suite for Advanced General Ensemble Simulations (SSAGES) that provides full GPU support for massively parallel applications of enhanced sampling methods such as adaptive biasing forces, harmonic bias, or forward flux sampling in the context of molecular dynamics simulations. By providing an intuitive interface that facilitates the management of a system's configuration, the inclusion of new collective variables, and the implementation of sophisticated free energy-based sampling methods, the PySAGES library serves as a general platform for the development and implementation of emerging simulation techniques. The capabilities, core features, and computational performance of this new tool are demonstrated with clear and concise examples pertaining to different classes of molecular systems. We anticipate that PySAGES will provide the scientific community with a robust and easily accessible platform to accelerate simulations, improve sampling, and enable facile estimation of free energies for a wide range of materials and processes.
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…
GPU Accelerated Simulation of Channeling Radiation of Relativistic Particles
在本文中，我们描述并演示了C++代码编写来确定粒子穿过取向单晶的轨迹和CUDA代码编写来评估带电粒子的辐射光谱与任意轨迹。CUDA/C++程序可以计算自旋0和1/2粒子的经典和量子力学辐射谱。我们包括多重库仑散射和能量损失，由于辐射发射产生的辐射光谱与实验光谱的正电子和电子。我们还演示了如何使用GPU来加速计算几个数量级。这将允许资金有限或很少使用超级计算机的研究小组进行数值计算，就好像它是一台超级计算机一样。我们发现，一个Titan V GPU可以取代多达100个并行运行的Xeon 36核CPU。我们还表明，为特定任务选择GPU将对性能产生很大影响，因为某些GPU具有比其他GPU更好的双精度性能。
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…
Differentiable and accelerated spherical harmonic and Wigner transforms
Mullard Space Science Laboratory, University College London, Dorking, RH5 6NT, UK（穆尔空间科学实验室，伦敦大学学院，多金，RH5 6NT，英国）；Alan Turing Institute, London, NW1 2DB, UK（艾伦·图灵研究所，伦敦，NW1 2DB，英国）
Maintaining computational load balance is important to the performant behavior of codes which operate under a distributed computing model. This is especially true for GPU architectures, which can suffer from memory oversubscription if improperly load balanced. We present enhancements to traditional load balancing approaches and explicitly target GPU architectures, exploring the resulting performance. A key component of our enhancements is the introduction of several GPU-amenable strategies for assessing compute work. These strategies are implemented and benchmarked to find the most optimal dat…
We assess the performance of the hybrid Open Accelerator (OpenACC) and Message Passing Interface (MPI) approach for multi-graphics processing units (GPUs) accelerated thermal lattice Boltzmann (LB) simulation. The OpenACC accelerates computation on a single GPU, and the MPI synchronizes the information between multiple GPUs. With a single GPU, the two-dimension (2D) simulation achieved 1.93 billion lattice updates per second (GLUPS) with a grid number of $8193^{2}$, and the three-dimension (3D) simulation achieved 1.04 GLUPS with a grid number of $385^{3}$, which is more than 76% of the theore…
We study large-scale applications using a GPU-accelerated version of the massively parallel Jülich universal quantum computer simulator (JUQCS--G). First, we benchmark JUWELS Booster, a GPU cluster with 3744 NVIDIA A100 Tensor Core GPUs. Then, we use JUQCS--G to study the relation between quantum annealing (QA) and the quantum approximate optimization algorithm (QAOA). We find that a very coarsely discretized version of QA, termed approximate quantum annealing (AQA), performs surprisingly well in comparison to the QAOA. It can either be used to initialize the QAOA, or to avoid the costly optim…
