As the engine for seismic imaging algorithms, stencil kernels modeling wave propagation are both computeand memoryintensive. This work targets improving the performance of wave equation based stencil code parallelized by OpenMP on a multi-core CPU. To achieve this goal, we explored two techniques: improving vectorization by using hardware SIMD technology, and reducing memory traffic to mitigate...

This paper explores stencil operations in CUDA to optimize on GPUs the Jacobi method for solving Laplace’s differential equation. The code keeps constant the access pattern through a large number of loop iterations, that way being representative of a wide set of iterative linear algebra algorithms. Optimizations are focused on data parallelism, threads deployment and the GPU memory hierarchy, w...

We present a series of optimization techniques for stencil computations on NVIDIA Kepler GPUs. Stencil computations with regular grids had been ported to the older generations of NVIDIA GPUs with significant performance improvements thanks to the higher memory bandwidth than conventional CPU-only systems. However, because of the architectural changes introduced with the latest generation of the...

The implementation of stencil computations on modern, massively parallel systems with GPUs and other accelerators currently relies on manually-tuned coding using low-level approaches like OpenCL and CUDA, which makes it a complex, time-consuming, and error-prone task. We describe how stencil computations can be programmed in our SkelCL approach that combines high level of programming abstractio...

Applic~atiori codes wliilblg ac,hievr performance f;lr Itw than tht, iidvertist~tl c,apabilities of cxistiug archit,tytnrcs, and this pmble~ii is worsening with irlc,reasingly-I)arallcl machines. For large-scale nunlorical ilpplic~at,iorls, stencil opcratioris oftm iiriposc the grcat,cr part of the wmputat,ional cost, ant1 the primary sources of incfficic3icy arc thr costs of lllrssagc passing ...

We investigate a set of adaptive±stencil, ®nite-volume schemes used to capture sharp fronts and shocks in a wide range of ®elds. Our objective is to determine the most promising methods available from this set for solving sharp-front advective±dispersive transport problems. Schemes are evaluated for a range of initial conditions, and for Peclet and Courant numbers. Based upon results from this ...

Stencil kernels arise in many scientific codes as the result from discretizing natural, continuous phenomenons. Many research works have designed stencil frameworks to help programmer optimize stencil kernels for performance, and to target CPUs or accelerators. However, existing stencil kernels, either library-based or languagebased necessitate to write distinct source codes for accelerated ker...

A stencil is a thin sheet of material, such as paper, plastic, or metal, with certain patterns cut from it. Applying a pigment through the cut-out holes produces a design on an underlying surface. Using multiple overlapping stencil layers, artists can create intricate, yet reproducible imagery on a variety of surfaces. Traditionally, artists have to design not only the final appearance, but als...

We derive a family of sixth-order compact finite-difference schemes for the three-dimensional Poisson’s equation. As opposed to other research regarding higher-order compact difference schemes, our approach includes consideration of the discretization of the source function on a compact finite-difference stencil. The schemes derived approximate the solution to Poisson’s equation on a compact st...

Auto-tuning Stencil Codes for Cache-Based Multicore Platforms by Kaushik Datta Doctor of Philosophy in Computer Science University of California, Berkeley Professor Katherine A. Yelick, Chair As clock frequencies have tapered off and the number of cores on a chip has taken off, the challenge of effectively utilizing these multicore systems has become increasingly important. However, the diversi...