High-Order Finite-differences on multi-threaded architectures using OCCA

TL;DR

This paper presents a portable, high-performance implementation of high-order finite-difference wave propagators on multi-core and many-core architectures using the OCCA runtime, with demonstrated efficiency on various hardware.

Contribution

It introduces thread parallel algorithms for high-order finite-difference wave propagators that are portable across architectures via OCCA, with performance evaluations included.

Findings

Achieved efficient performance on multiple architectures.

Demonstrated portability through OCCA interface.

Validated with synthetic test case results.

Abstract

High-order finite-difference methods are commonly used in wave propagators for industrial subsurface imaging algorithms. Computational aspects of the reduced linear elastic vertical transversely isotropic propagator are considered. Thread parallel algorithms suitable for implementing this propagator on multi-core and many-core processing devices are introduced. Portability is addressed through the use of the \OCCA runtime programming interface. Finally, performance results are shown for various architectures on a representative synthetic test case.