Generating Device-specific GPU Code for Local Operators in Medical Imaging

Membarth R, Hannig F, Teich J, Körner M, Eckert W (2012)

Publication Type: Conference contribution

Publication year: 2012

Publisher: IEEE Press

Edited Volumes: Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium, IPDPS 2012

City/Town: New York, NY, USA

Pages Range: 569-581

Conference Proceedings Title: Proc. of the 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS)

Event location: Shanghai CN

ISBN: 978-1-4673-0975-2

DOI: 10.1109/IPDPS.2012.59

Abstract

To cope with the complexity of programming GPU accelerators for medical imaging computations, we developed a framework to describe image processing kernels in a domain-specific language, which is embedded into C++. The description uses decoupled access/execute metadata, which allow the programmer to specify both execution constraints and memory access patterns of kernels. A source-to-source compiler translates this high-level description into low-level CUDA and Open CL code with automatic support for boundary handling and filter masks. Taking the annotated metadata and the characteristics of the parallel GPU execution model into account, two-layered parallel implementations - utilizing SPMD and MPMD parallelism - are generated. An abstract hardware model of graphics card architectures allows to model GPUs of multiple vendors like AMD and NVIDIA, and to generate device-specific code for multiple targets. It is shown that the generated code is faster than manual implementations and those relying on hardware support for boundary handling. Implementations from Rapid Mind, a commercial framework for GPU programming, are outperformed and similar results achieved compared to the GPU backend of the widely used image processing library Open CV. © 2012 IEEE.

Authors with CRIS profile

Richard Membarth Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design) Frank Hannig Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design) Jürgen Teich Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)

Related research project(s)

Multi-core Architectures and Programming Jan. 1, 2009 - Dec. 31, 2014 A Domain-Specific Language and GPU Target Code Generator for Image Processing Applications (HIPAcc) Jan. 1, 2009 - Jan. 1, 2012

How to cite

APA:

Membarth, R., Hannig, F., Teich, J., Körner, M., & Eckert, W. (2012). Generating Device-specific GPU Code for Local Operators in Medical Imaging. In Proc. of the 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS) (pp. 569-581). Shanghai, CN: New York, NY, USA: IEEE Press.

MLA:

Membarth, Richard, et al. "Generating Device-specific GPU Code for Local Operators in Medical Imaging." Proceedings of the 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS), Shanghai New York, NY, USA: IEEE Press, 2012. 569-581.

BibTeX: Download