# Investigating the OPS intermediate representation to target GPUs in the Devito DSL

## TL;DR

This paper explores integrating the OPS API into the Devito DSL to enable efficient GPU execution of finite difference PDE solutions, resulting in significant performance improvements.

## Contribution

It introduces an OPS backend for Devito, facilitating GPU targeting and demonstrating notable speedups over the native Devito implementation.

## Key findings

- Significant speedups achieved on GPU compared to native Devito.
- Successful integration of OPS API into Devito for GPU support.
- Enhanced performance for seismic inversion PDE solutions.

## Abstract

The Devito DSL is a code generation tool for the solution of partial differential equations using the finite difference method specifically aimed at seismic inversion problems. In this work we investigate the integration of OPS, an API to generate highly optimized code for applications running on structured meshes targeting various platforms, within Devito as a mean of bringing it to the GPU realm by providing an implementation of a OPS backend in Devito, obtaining considerable speed ups compared to the core Devito backend.