Chengwen Zhong

Implementation and Analysis of AES Encryption on GPU

GPU is continuing its trend of vastly outperforming CPU while becoming more general purpose. In order to improve the efficiency of AES algorithm, this paper proposed a CUDA implementation of Electronic Codebook (ECB) mode encoding process and Cipher Feedback (CBC) mode decoding process on GPU. In our implementation, the frequently accessed T-boxes were allocated on on-chip shared memory and the granularity that one thread handles a 16 Bytes AES block was adopted. Finally, we achieved the highest performance of around 60 Gbps throughput on NVIDIA Tesla C2050 GPU, which runs up to 50 times faster than a sequential implementation based on Intel Core i7-920 2.66GHz CPU. In addition, we discussed the optimization under some practical application scenarios such as overlapping GPU processing and data transfer.

Implicit unified gas-kinetic scheme for steady state solutions in all flow regimes

This paper presents an implicit unified gas-kinetic scheme (UGKS) for non-equilibrium steady state flow computation. The UGKS is a direct modeling method for flow simulation in all regimes with the updates of both macroscopic flow variables and microscopic gas distribution function. By solving the macroscopic equations implicitly, a predicted equilibrium state can be obtained first through iterations. With the newly predicted equilibrium state, the evolution equation of the gas distribution function and the corresponding collision term can be discretized in a fully implicit way for fast convergence through iterations as well. The lower-upper symmetric Gauss-Seidel (LU-SGS) factorization method is implemented to solve both macroscopic and microscopic equations, which improves the efficiency of the scheme. Since the UGKS is a direct modeling method and its physical solution depends on the mesh resolution and the local time step, a physical time step needs to be fixed before using an implicit iterative technique with a pseudo-time marching step. Therefore, the physical time step in the current implicit scheme is determined by the same way as that in the explicit UGKS for capturing the physical solution in all flow regimes, but the convergence to a steady state speeds up through the adoption of a numerical time step with large CFL number. Many numerical test cases in different flow regimes from low speed to hypersonic ones, such as the Couette flow, cavity flow, and the flow passing over a cylinder, are computed to validate the current implicit method. The overall efficiency of the implicit UGKS can be improved by one or two orders of magnitude in comparison with the explicit one.

Unified gas-kinetic scheme for diatomic molecular simulations in all flow regimes

A unified gas-kinetic scheme (UGKS) is constructed for both continuum and rarefied flow computations. The underlying principle for the development of UGKS is the direct modeling for the gas evolution process from the kinetic to the hydrodynamic scale, which is used in the flux construction across a cell interface. More specifically, the physical process from the kinetic particle free transport to the hydrodynamic pressure wave propagation is recovered in the flux function. In the previous study, the UGKS has been developed mainly for monatomic gas with particle translational motion only. The construction of time evolution solution is based on the BGK, Shakhov, and ES-BGK models. The UGKS has been validated through extensive numerical tests. In this paper, a UGKS for diatomic gas will be constructed, where the gas-kinetic Rykov model with a Landau-Teller-Jeans-type rotational energy relaxation is used in the numerical scheme. The new scheme will be tested in many cases, such as homogeneous flow relaxation, shock structure calculations, hypersonic flow passing a flat plate, and the flow around a blunt circular cylinder. The analytic, DSMC, and experimental measurements will be used for validating the solutions of UGKS.

Non-body-fitted Cartesian-mesh simulation of highly turbulent flows using multi-relaxation-time lattice Boltzmann method

This paper presents a lattice Boltzmann method (LBM) based study aimed at numerical simulation of highly turbulent and largely inclined flow around obstacles of curved geometry using non-body-fitted Cartesian meshes. The approach features (1) combining the interpolated bounce-back scheme with the LBM of multi-relaxation-time (MRT) type to enable the use of simple Cartesian mesh for the flow cases even with complex geometries; and (2) incorporating the Spalart-Allmaras (SA) turbulence model into LBM in order to represent the turbulent flow effect. The numerical experiments are performed corresponding to flows around an NACA0012 airfoil at Re=5x10^5 and around a flat plate at Re=2x10^4, respectively. The agreement between all simulation results obtained from this study and the data provided by other literature demonstrates the reliability of the enhanced LBM proposed in this paper for simulating, simply on Cartesian meshes, complex flows that may involve bodies of curved boundary, high Reynolds number, and large angle of attack.

Implementation of dual time-stepping strategy of the gas-kinetic scheme for unsteady flow simulations

In our study, the dual time-stepping strategy of the gas-kinetic scheme is constructed and used for the simulation of unsteady flows. In comparison to the previous implicit gas-kinetic scheme, both the inviscid and viscous flux Jacobian are considered in our work, and the linear system of the pseudo-steady-state is solved by applying generalized minimal residual algorithm. The accuracy is validated by several numerical cases, the incompressible flow around blunt bodies (stationary circular cylinder and square cylinder), and the transonic buffet on the NACA0012 airfoil under hybrid mesh. The numerical cases also demonstrate that the present method is applicable to approach the fluid flows from laminar to turbulent and from incompressible to compressible. Finally, the case of acoustic pressure pulse is carried out to evaluate the effects of enlarged time step, and the side effect of enlarged time step is explained. Compared with the explicit gas-kinetic scheme, the proposed scheme can greatly accelerate the computation and reduce the computational costs for unsteady flow simulations.

Study of Web Information Extraction and Classification Method

In the vast information on the Internet, it is now the key problem that how to download information we need most quickly and expediently, to keep the information in database, and organize it for intending usage. Nowadays, searching engine aims at the different needs of different customers, providing an individualization service, for checking the rate of comprehensions rather than the rate of accuracy. If making it by manual duplicate and stick, we can imagine how the efficiency and cost would be. However, Web information extraction and classification aim to resolve this problem. In this paper, it is discussed that a method how to complete to follow the appointed Web site or Web page according to the users request using the technique of ASP, carry on analysis, comparison, sample, and data storage, extract to obtain the information we need from the Web, and organize organically the information adopting classification technique with the Web page contents to provide for later use.

Filter-matrix lattice Boltzmann simulation of lid-driven deep-cavity flows, Part I - Steady flows

This paper seeks to make a systematic study over a series of lid-driven flow in various deep cavities using the filter-matrix lattice Boltzmann (FMLB) model. A concise description of the FMLB model is presented in this paper, and important numerical considerations for effective use of the FMLB model are also clearly elucidated. In particular, the selection of a free parameter employed to appropriately control the weight of the third-order terms in the FMLB solution vector is carefully examined, resulting in some general suggestions that may render the FMLB stability consistently secured for simulations of different cavity flow scenarios. Employing the FMLB and the lattice Bhatnagar-Gross-Krook (LBGK) methods for comparison purpose, the first series of test cases correspond to the lid-driven cavity flows with a low Reynolds number (Re=0.01) at a variety of aspect ratios; the simulation results demonstrate that the FMLB model is superior to the LBGK method in terms of numerical stability and, particularly, the FMLB result can reach quite good agreement with the benchmark solution even if the aspect ratio goes up to 15. Then, the FMLB model is used to compute the steady flows for deep cavities with aspect ratios ranging from 1.5 to 7 and elevated Reynolds numbers ranging from 100 to 5000; a number of features of steady flows, such as the locations, strengths, and sizes of the vortices, as well as the effects of Reynolds number and aspect ratio on the vortex structure, are all predicted by the FMLB model with an obviously improved accuracy when compared to some other available numerical results.

Filter-matrix lattice Boltzmann simulation of lid-driven deep-cavity flows, Part II - Flow bifurcation

Following the first part of this study, the filter-matrix lattice Boltzmann (FMLB) model is now applied to the investigation of the bifurcation behavior in the lid-driven deep-cavity flow. In this second part, the first Hopf bifurcations in the lid-driven cavity flow patterns with aspect ratios of 1-5 are examined in detail, revealing that the critical Reynolds number converges to a constant value with the increase of the cavity depth, and that the time-dependent vortex structures are periodic or quasi-periodic once this critical Reynolds number is exceeded. Through comparison against the relevant numerical results reported in the available literature, the present FMLB approach demonstrates its effectiveness and usefulness in studying the bifurcation phenomena arising in complex lid-driven deep-cavity flows.

A parallel lattice Boltzmann method for large eddy simulation on multiple GPUs

To improve the simulation efficiency of turbulent fluid flows at high Reynolds numbers with large eddy dynamics, a CUDA-based simulation solution of lattice Boltzmann method for large eddy simulation (LES) using multiple graphics processing units (GPUs) is proposed. Our solution adopts the “collision after propagation” lattice evolution way and puts the misaligned propagation phase at global memory read process. The latest GPU platform allows a single CPU thread to control up to four GPUs that run in parallel. In order to make use of multiple GPUs, the whole working set is evenly partitioned into sub-domains. We implement Smagorinsky model and Vreman model respectively to verify our multi-GPU solution. These two LES models have different relaxation time calculation behavior and lead to different CUDA implementation characteristics. The implementation based on Smagorinsky model achieves 190 times speedup over the sequential implementation on CPU, while the implementation based on Vreman model archives more than 90 times speedup. The experimental results show that the parallel performance of our multi-GPU solution scales very well on multiple GPUs. Therefore large-scale (up to 10,240