CUDACLAW: A high-performance programmable GPU framework for the solution of hyperbolic PDEs

TL;DR

cudaclaw is a GPU framework that enables scientists to efficiently solve hyperbolic PDEs without requiring CUDA programming expertise, optimizing data transfer and computation for high performance.

Contribution

It introduces a CUDA-based framework that simplifies solving hyperbolic PDEs on GPUs, handling data management and parallel execution automatically.

Findings

Achieves high computational throughput on GPUs.

Reduces user burden by abstracting CUDA details.

Includes optimizations for memory access efficiency.

Abstract

We present cudaclaw, a CUDA-based high performance data-parallel framework for the solution of multidimensional hyperbolic partial differential equation (PDE) systems, equations describing wave motion. cudaclaw allows computational scientists to solve such systems on GPUs without being burdened by the need to write CUDA code, worry about thread and block details, data layout, and data movement between the different levels of the memory hierarchy. The user defines the set of PDEs to be solved via a CUDA- independent serial Riemann solver and the framework takes care of orchestrating the computations and data transfers to maximize arithmetic throughput. cudaclaw treats the different spatial dimensions separately to allow suitable block sizes and dimensions to be used in the different directions, and includes a number of optimizations to minimize access to global memory.