Data structure and movement for lattice-based simulations

A GPU accelerated study of aqueous humor dynamics in human eyes using the lattice Boltzmann method

In this paper, we presented a 3D human eyes aqueous humor (AH) dynamics model, and additionally, designed and optimized it using GPU technology. First, the feasibility of the model is demonstrated through validation. Then, the effect of different factors on AH flow was investigated using the validated model. The experimental results showed that AH flow more rapidly when standing than supine; the intraocular temperature has the greatest effect on AH flow compared to other factors; the AH secretion rate and trabecular meshwork (TM) permeability had a greater effect on intraocular pressure (IOP). Corneal indentation and ovoid anterior chamber (AC) can also affect AH flow. Finally, the PartSparse algorithm based GPU can save more than 50% of the memory consumption and achieves a performance of 1491.29 MLUPS and a Speedup of 837.61 times.

Propagation pattern for moment representation of the lattice Boltzmann method

A propagation pattern for the moment representation of the regularized lattice Boltzmann method (LBM) in three dimensions is presented. Using effectively lossless compression, the simulation state is stored as a set of moments of the lattice Boltzmann distribution function, instead of the distribution function itself. An efficient cache-aware propagation pattern for this moment representation has the effect of substantially reducing both the storage and memory bandwidth required for LBM simulations. This paper extends recent work with the moment representation by expanding the performance analysis on central processing unit (CPU) architectures, considering how boundary conditions are implemented, and demonstrating the effectiveness of the moment representation on a graphics processing unit (GPU) architecture.

Delayed difference scheme for large scale scientific simulations

We argue that the current heterogeneous computing environment mimics a complex nonlinear system which needs to borrow the concept of time-scale separation and the delayed difference approach from statistical mechanics and nonlinear dynamics. We show that by replacing the usual difference equations approach by a delayed difference equations approach, the sequential fraction of many scientific computing algorithms can be substantially reduced. We also provide a comprehensive theoretical analysis to establish that the error and stability of our scheme is of the same order as existing schemes for a large, well-characterized class of problems.