Parallel implementation features of self-gravitating gas dynamics modeling on multiple GPUs are considered applying the GPU-Direct technology. The parallel algorithm for solving of the self-gravitating gas dynamics problem based on hybrid OpenMP-CUDA parallel programming model has been described in detail. The gas-dynamic forces are calculated by the modified SPH-method (Smoothed Particle Hydrodynamics) while the N-body problem gravitational interaction is obtained by the direct method (so-called Particle-Particle algorithm). The key factor in the SPH-method performance is creation of the neighbor lists of the particles which contribute into the gas-dynamic forces calculation. Our implementation is based on hierarchical grid sorting method using a cascading algorithm for parallel computations of partial sums at CUDA block. The parallelization efficiency of the algorithm for various GPUs of the Nvidia Tesla line (K20, K40, K80) is studied in the framework of galactic’ gaseous halos collisions models by the SPH-method.

Smoothed-Particle Hydrodynamics Models: Implementation Features on GPUs

Навстречу юбилею: итоги аккредитации Волгоградского государственного университета

Первые результаты нового этапа развития учебно-научной работы в рамках деятельности класса Конфуция Новосибирского государственного университета

О подготовке научно-педагогических кадров в аспирантуре Волгоградского государственного университета

The article provides detailed comments to interpretation of concepts of a «school of thought», requirements to sections «Basic Problems of scientific school» and «Criteria of scientific and teaching teams», applying for a status of a school of thought», forming a basis of «The Norms of school of thoughts» of numbers of State Universities of railway engineering in relation to the school of thought «Dynamics and strength of units and equipments of transport systems» at the Samara State Transport University, which was created and managed by the article’s author for more than 10 years.

Цель, задачи и критерии формирования и развития научной школы «Динамика и прочность агрегатов и оборудования транспортных систем» Самарского государственного университета путей сообщения

In the paper we discuss the main features of the software package for numerical simulations of the surface water dynamics. We consider an approximation of the shallow water equations together with the parallel technologies for NVIDIA CUDA graphics processors. The numerical hydrodynamic code is based on the combined Lagrangian-Euler method (CSPH-TVD). We focused on the features of the parallel implementation of Tesla line of graphics processors: C2070, K20, K40, K80. By using hierarchical grid systems at different spatial scales we increase the efficiency of the computing resources usage and speed up our simulations of a various flooding problems.

/pdf/numerical-model-of-shallow-water-the-use-of-nvidia-cuda-cfkoq6aq6r.pdf

Numerical Model of Shallow Water: The Use of NVIDIA CUDA Graphics Processors

In the paper we discuss the main features of the software package for numerical simulations of the surface water dynamics. We consider an approximation of the shallow water equations together with the parallel technologies for NVIDIA CUDA graphics processors. The numerical hydrodynamic code is based on the combined Lagrangian-Euler method~(CSPH-TVD). We focused on the features of the parallel implementation of Tesla line of graphics processors: C2070, K20, K40, K80. By using hierarchical grid systems at different spatial scales we increase the efficiency of the computing resources usage and speed up our simulations of a various flooding problems.

Numerical Model of Shallow Water: the Use of NVIDIA CUDA Graphics Processors

The problem of choice of boundary conditions are discussed for the case of numerical integration of the shallow water equations on a substantially irregular relief. In modeling of unsteady surface water flows has a dynamic boundary partitioning liquid and dry bottom. The situation is complicated by the emergence of sub- and supercritical flow regimes for the problems of seasonal floodplain flooding, flash floods, tsunami landfalls. Analysis of the use of various methods of setting conditions for the physical quantities of liquid when the settlement of the boundary shows the advantages of using the waterfall type conditions in the presence of strong inhomogeneities landforms. When there is a waterfall on the border of the computational domain and heterogeneity of the relief in the vicinity of the boundary portion may occur, which is formed by the region of critical flow with the formation of a hydraulic jump, which greatly weakens the effect of the waterfall on the flow pattern upstream.

The problem of boundary conditions for the shallow water equations (Russian)

The correct description of the surface water dynamics in the model of shallow water requires accounting for friction. To simulate a channel flow in the Chezy model the constant Manning roughness coefficient is frequently used. The Manning coefficient n M is an integral parameter which accounts for a large number of physical factors determining the flow braking. We used computational simulations in a shallow water model to determine the relationship between the Manning coefficient and the parameters of small-scale perturbations of a bottom in a long channel. Comparing the transverse water velocity profiles in the channel obtained in the models with a perturbed bottom without bottom friction and with bottom friction on a smooth bottom, we constructed the dependence of n M on the amplitude and spatial scale of perturbation of the bottom relief.

Bottom friction models for shallow water equations: Manning's roughness coefficient and small-scale bottom heterogeneity

We consider the possibilities of numerical hydrodynamic modeling for the construction of cadastral maps of dangerous flood zones on the example of the Volgograd region. The numerical model is based on shallow water equations and Combined Smooth Particle Hydrodynamics – Total Variation Diminishing (CSPH-TVD) method for integrating hyperbolic differential equations. To create a digital elevation model (DEM), we use an iterative approach, assimilating all available spatial data, including remote sensing data, topographic measurements and observations of flooding in different years. The agreement of the observed data with the numerical simulation results can significantly improve the quality of the DEM. Digital elevation models are based on elevation matrices in increments of 3 to 10 meters, depending on the topography of the study area. The result of our work is a set of maps with flood boundaries for different probabilities of catastrophic events. For each such flood, we calculate the river hydrograph and its integral characteristics.

/pdf/creation-of-cadastral-maps-of-flooding-based-on-numerical-33sw0v1v08.pdf

Tatyana Dyakonova

Papers

Numerical Model of Shallow Water: The Use of NVIDIA CUDA Graphics Processors

Numerical Model of Shallow Water: the Use of NVIDIA CUDA Graphics Processors

The problem of boundary conditions for the shallow water equations (Russian)

Bottom friction models for shallow water equations: Manning's roughness coefficient and small-scale bottom heterogeneity

Creation of Cadastral Maps of Flooding Based on Numerical Modeling