A Domain-Specific Language and Compiler for Stencil Computations on Short-Vector SIMD and GPU Architectures
Ohio State University
Compilers for Parallel Computing Workshop (CPC), 2013
@article{henretty2013domain,
title={A Domain-Specific Language and Compiler for Stencil Computations on Short-Vector SIMD and GPU Architectures},
author={Henretty, Tom and Holewinski, Justin and Veras, Richard and Franchetti, Franz and Pouchet, Louis-No{"e}l and Ramanujam, J and Rountev, Atanas and Sadayappan, P},
year={2013}
}
Stencil computations are an integral part of applications in a number of scientific computing domains, such as image processing and partial differential equations. We describe a domain-specific language for regular stencil computations, that allows specification of the computations in a concise manner. We describe a multi-target compiler for this DSL, that generates optimized code for multi-core processors with short-vector SIMD engines, as well as GPUs. The hardware differences between these two types of architecture prompt different optimization strategies for the compiler. A data layout transformation along with split tiling is used for multi-core CPUs, while overlapped tiling is used for GPUs. We evaluate our domain-specific compiler for a number of benchmarks on CPU and GPU platforms.
August 19, 2013 by hgpu
Recent source codes
SYCL-BLAS: An implementation of BLAS using the SYCL open standard for acceleration on OpenCL devices
