3D finite difference computation on GPUs using CUDA

By Paulius Micikevicius. GPGPU-2, 2009

611 citations recorded by the source corpus.

Subfields: Geophysics, Ocean Engineering

View Paper: https://www.semanticscholar.org/paper/e104e4e93e84ff23cb48d832417879b18409c18d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Geophysics

In Ocean Engineering

Citations in the corpus

The pochoir stencil compiler (2011, methodology): “Figure 3 shows Pochoir’s performance on a wider range of benchmarks, including heat equation (Heat) [13] on a 2D grid, a 2D torus, and a 4D grid; Conway’s game of Life (Life) [18]; 3D finite-difference wave equation (Wave) [32]; lattice Boltzmann method (LBM) [30]; RNA secondary structure prediction”
High-order finite-element seismic wave propagation modeling with MPI on a large GPU cluster (2010, methodology): “For a broad range of applications, depending on the algorithms used and the ease of parallelization, applications ported to the GPU in the literature have achieved speedups that typically range from 3x to 50x with respect to calculations run on a single CPU core.”
3.5-D Blocking Optimization for Stencil Computations on Modern CPUs and GPUs (2010, methodology): “Recently, there are a number of publications that map SC and LBM to graphics processing units (GPU) [15], [24], [25], [26], [27], [28], [29].”
Forward and adjoint simulations of seismic wave propagation on fully unstructured hexahedral meshes (2011, methodology): “Once the outer elements have been identified following a standard procedure (see e.g. Danielson & Namburu 1998; Martin et al. 2008a; Micikevicius 2009; Michéa & Komatitsch 2010; Komatitsch 2010a, 2011), MPI buffers are filled and a non-blocking MPI call is issued, which initiates communication and…”
Accelerating Wright–Fisher Forward Simulations on the Graphics Processing Unit (2016, positive): “Beyond GO Fish, solving Wright – Fisher diffusion equations in programs like d a d i (Gutenkunst et al. 2009) can likewise be sped up through parallelization on the GPU (Lions et al. 2001; Komatitsch et al. 2009; Micikevicius 2009; Tutkun and Edis 2012).”
Spatially Parallel Convolutions (2018, background): “To a degree, the problem of working with spatial tensors too large to fit on a single GPU is a solved problem (Micikevicius, 2009; Coates et al., 2013).”
PATUS: A Code Generation and Autotuning Framework for Parallel Iterative Stencil Computations on Modern Microarchitectures (2011, background): “Recent performance studies for stencil computations on a variety of platforms using autotuning include [15]; [16] gives results on stencil computations on a GPU.”
State-of-the-art in heterogeneous computing (2010, background): “cling (GPU) [118]; Domain blocking, loop unrolling (CBEA) [12]; Auto-tuning (CBEA, GPU) [46]; Time skewing”