Devito: Towards a generic Finite Difference DSL using Symbolic Python. Michael Lange, Navjot Kukreja, Mathias Louboutin, Fabio Luporini, Felippe Vieira, Vincenzo Pandolfo, Paulius Velesko, Paulius Kazakas, & Gerard Gorman pyHPC 2016 conference, September, 2016. GitHub repository: https://github.com/opesci/devito
Devito: Towards a generic Finite Difference DSL using Symbolic Python [link]Paper  abstract   bibtex   
Domain specific languages (DSL) have been used in a variety of fields to express complex scientific problems in a concise manner and provide automated performance optimization for a range of computational architectures. As such DSLs provide a powerful mechanism to speed up scientific Python computation that goes beyond traditional vectorization and pre-compilation approaches, while allowing domain scientists to build applications within the comforts of the Python software ecosystem. In this paper we present Devito, a new finite difference DSL that provides optimized stencil computation from high-level problem specifications based on symbolic Python expressions. We demonstrate Devito's symbolic API and performance advantages over traditional Python acceleration methods before highlighting its use in the scientific context of seismic inversion problems.
@article{michael_lange_devito:_2016,
	title = {Devito: {Towards} a generic {Finite} {Difference} {DSL} using {Symbolic} {Python}},
	url = {http://arxiv.org/abs/1609.03361},
	abstract = {Domain specific languages (DSL) have been used in a variety of fields to express complex scientific problems in a concise manner and provide automated performance optimization for a range of computational architectures. As such DSLs provide a powerful mechanism to speed up scientific Python computation that goes beyond traditional vectorization and pre-compilation approaches, while allowing domain scientists to build applications within the comforts of the Python software ecosystem. In this paper we present Devito, a new finite difference DSL that provides optimized stencil computation from high-level problem specifications based on symbolic Python expressions. We demonstrate Devito's symbolic API and performance advantages over traditional Python acceleration methods before highlighting its use in the scientific context of seismic inversion problems.},
	journal = {pyHPC 2016 conference},
	author = {{Michael Lange} and {Navjot Kukreja} and {Mathias Louboutin} and {Fabio Luporini} and {Felippe Vieira} and {Vincenzo Pandolfo} and {Paulius Velesko} and {Paulius Kazakas} and {Gerard Gorman}},
	month = sep,
	year = {2016},
	note = {GitHub repository: https://github.com/opesci/devito},
	keywords = {JIT, Just-in-time, finite difference method, numerical analysis, uses sympy},
}
Downloads: 0