Optimised Finite Difference Computation from Symbolic Equations | SciPy 2017 | Michael Lange

Domain-specific high-productivity environments are playing an increasingly important role in scientific computing due to the increased levels of abstraction and automation they provide. In this talk we introduce Devito, an open-source domain-specific framework for solving partial differential equations from symbolic problem definitions by the finite difference method. We highlight the generation and automated execution of highly optimized stencil code from only a few lines of high-level symbolic Python for a set of operators used in seismic inversion problems, before exploring the use of Devito for a range of scientific equations.