遇见数据集

Nonhomogeneous_Turbulence_Sim_Profiling

收藏
Zenodo2026-08-13 更新2026-08-20 收录
官方服务:

资源简介:

SAND No. SAND2026-25072C Contains data from simulation of non-homogeneous turbulent flows with PDF methods. Stored in a hdf5 format, the data is organized by device (CPU, GPU, Loihi), method of simulation (Eulerian, Lagrangian, or mixed), and number of random walks. The data includes an array of random walk data (for Eulerian or mixed, the array is number of mesh-points by number of timesteps in size, while Lagrangian is number of walks by number of timesteps in size) as well as the starting state for each of 10 trials. For Eulerian/mixed, these arrays hold the number of walkers at each mesh-point in discretized velocity space at each timestep. For Lagrangian, these arrays hold the current velocity of each walker at each timestep. Each hdf5 dataset also has attributes related to the simulation set-up, including the bounds of the velocity space (L), the discretization of velocity space (ds), the discretization of time (dt), the number of connected “neighbor” mesh-points, the number of simulated parameter sets (num_samples), the number of timesteps per parameter set (steps_per_update), the mean parameters (mean_1), the variance parameters (var_1), the characteristic timescale parameters (TL_1), the recorded execution time (total_time), and the recorded energy consumption (total_energy). Sample Data Set: dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘L’] = 3.0dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘ds’] = 0.05dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘dt’] = 10dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘neighbors’] = 15dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘num_samples’] = 100dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘steps_per_update’] = 100 dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘total_time’] = 11.563927999999974dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘trial_v1’].attrs[‘total_energy’] = 23.066640107272022 dataset[‘loihi’][‘eulerian’][‘50000_walkers’][‘probes_off’][‘start_state_v1’][:] = [ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 0 2 1 3 1 7 8 7 15 19 32 29 36 61 75 82 83 117 154 190 219 267 309 356 424 505 586 594 672 793 926 1015 1110 1264 1268 1307 1418 1373 1562 1659 1590 1661 1767 1802 1693 1713 1694 1660 1559 1516 1475 1352 1323 1178 1168 1027 939 853 802 725 590 554 433 375 332 310 238 218 176 143 121 91 75 72 41 42 28 22 19 19 11 13 8 4 2 6 3 2 0 2 0 0 0 1 0 0 0 0 0 0]

提供机构:
Zenodo
创建时间:
2026-08-13
二维码
社区交流群
二维码
科研交流群
商业服务