Publikation

Target-Specific Refinement of Multigrid Codes

Richard Membarth, Philipp Slusallek, Marcel Köster, Roland Leißa, Sebastian Hack

In: Proceedings of the Fourth International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing (WOLFHPC). International Workshop on Domain-Specific Languages and High-Level Frameworks for High Performance Computing (WOLFHPC-2014) befindet sich SC14: The International Conference for High Performance Computing, Networking, Storage, and Analysis November 17-17 New Orleans LA United States Seiten 52-57 IEEE 11/2014.

Abstrakt

This paper applies partial evaluation to stage a stencil code Domain-Specific Language (DSL) onto a functional and imperative programming language. Platform-specific primitives such as scheduling or vectorization, and algorithmic variants such as boundary handling are factored out into a library that make up the elements of that DSL. We show how partial evaluation can eliminate all overhead of this separation of concerns and creates code that resembles hand- crafted versions for a particular target platform. We evaluate our technique by implementing a DSL for the V-cycle multigrid iteration. Our approach generates code for AMD and NVIDIA GPUs (via SPIR and NVVM) as well as for CPUs using AVX/AVX2 alike from the same high-level DSL program. First results show that we achieve a speedup of up to 3× on the CPU by vectorizing multigrid components and a speedup of up to 2× on the GPU by merging the computation of multigrid components.

Weitere Links

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence