Automatic Optimization of In-Flight Memory Transactions for GPU Accelerators based on a Domain-Specific Language for Medical Imaging

Beitrag bei einer Tagung


Details zur Publikation

Autor(en): Membarth R, Hannig F, Teich J, Körner M, Eckert W
Titel Sammelwerk: Proceedings - 2012 11th International Symposium on Parallel and Distributed Computing, ISPDC 2012
Verlag: IEEE Press
Verlagsort: New York, NY, USA
Jahr der Veröffentlichung: 2012
Tagungsband: Proc. of the 11th International Symposium on Parallel and Distributed Computing (ISPDC)
Seitenbereich: 211-218
ISBN: 978-1-4673-2599-8


Abstract


An efficient memory bandwidth utilization for GPU accelerators is crucial for memory bound applications. In medical imaging, the performance of many kernels is limited by the available memory bandwidth since only a few operations are performed per pixel. For such kernels only a fraction of the compute power provided by GPU accelerators can be exploited and performance is predetermined by memory bandwidth. As a remedy, this paper investigates the optimal utilization of available memory bandwidth by means of increasing in-flight memory transactions. Instead of doing this manually for different GPU accelerators, the required CUDA and OpenCL code is automatically generated from descriptions in a Domain-Specific Language (DSL) for the considered application domain. Moreover, the DSL is extended to also support global reduction operators. We show that the generated target-specific code improves bandwidth utilization for memory-bound kernels significantly. Moreover, competitive performance compared to the GPU back end of the widely used image processing library OpenCV can be achieved. © 2012 IEEE.



FAU-Autoren / FAU-Herausgeber

Hannig, Frank PD Dr.-Ing.
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)
Membarth, Richard
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)
Teich, Jürgen Prof. Dr.-Ing.
Lehrstuhl für Informatik 12 (Hardware-Software-Co-Design)


Autor(en) der externen Einrichtung(en)
Siemens AG


Zitierweisen

APA:
Membarth, R., Hannig, F., Teich, J., Körner, M., & Eckert, W. (2012). Automatic Optimization of In-Flight Memory Transactions for GPU Accelerators based on a Domain-Specific Language for Medical Imaging. In Proc. of the 11th International Symposium on Parallel and Distributed Computing (ISPDC) (pp. 211-218). Munich, DE: New York, NY, USA: IEEE Press.

MLA:
Membarth, Richard, et al. "Automatic Optimization of In-Flight Memory Transactions for GPU Accelerators based on a Domain-Specific Language for Medical Imaging." Proceedings of the 11th International Symposium on Parallel and Distributed Computing (ISPDC), Munich New York, NY, USA: IEEE Press, 2012. 211-218.

BibTeX: 

Zuletzt aktualisiert 2018-09-08 um 23:08