Accelerator processors allow energy-efficient computation at
high performance, especially for computationintensive
applications. There exists a plethora of different accelerator
architectures, such as GPUs and the Cell Broadband Engine. Each
accelerator has its own programming language, but the recently
introduced OpenCL language unifies accelerator programming
languages. Hereby, OpenCL achieves functional protability,
allowing to reduce the development time of kernels. Functional
portability however has limited value without performance
portability: the possibility to re-use optimized kernels with
good performance. This paper investigates the specificity of
code optimizations to accelerator architecture and the severity
of lack of performance portability.
%0 Conference Paper
%1 Rul2010-si
%A Rul, Sean
%A Vandierendonck, Hans
%A D'Haene, Joris
%A De Bosschere, Koen
%B 2010 Symposium on Application Accelerators in High Performance
Computing (SAAHPC '10)
%D 2010
%I biblio.ugent.be
%K Expose OpenCL Performance_portability
%T An experimental study on performance portability of OpenCL
kernels
%X Accelerator processors allow energy-efficient computation at
high performance, especially for computationintensive
applications. There exists a plethora of different accelerator
architectures, such as GPUs and the Cell Broadband Engine. Each
accelerator has its own programming language, but the recently
introduced OpenCL language unifies accelerator programming
languages. Hereby, OpenCL achieves functional protability,
allowing to reduce the development time of kernels. Functional
portability however has limited value without performance
portability: the possibility to re-use optimized kernels with
good performance. This paper investigates the specificity of
code optimizations to accelerator architecture and the severity
of lack of performance portability.
@inproceedings{Rul2010-si,
abstract = {Accelerator processors allow energy-efficient computation at
high performance, especially for computationintensive
applications. There exists a plethora of different accelerator
architectures, such as GPUs and the Cell Broadband Engine. Each
accelerator has its own programming language, but the recently
introduced OpenCL language unifies accelerator programming
languages. Hereby, OpenCL achieves functional protability,
allowing to reduce the development time of kernels. Functional
portability however has limited value without performance
portability: the possibility to re-use optimized kernels with
good performance. This paper investigates the specificity of
code optimizations to accelerator architecture and the severity
of lack of performance portability.},
added-at = {2015-04-10T18:02:47.000+0200},
author = {Rul, Sean and Vandierendonck, Hans and D'Haene, Joris and De Bosschere, Koen},
biburl = {https://www.bibsonomy.org/bibtex/24d9ad5ff919fc6eebc6c7a9af4fe5f68/christophv},
booktitle = {2010 Symposium on Application Accelerators in High Performance
Computing ({SAAHPC} '10)},
interhash = {898495a93358f4fe8e339a9f0b7af805},
intrahash = {4d9ad5ff919fc6eebc6c7a9af4fe5f68},
keywords = {Expose OpenCL Performance_portability},
publisher = {biblio.ugent.be},
timestamp = {2016-01-04T14:22:08.000+0100},
title = {An experimental study on performance portability of {OpenCL}
kernels},
year = 2010
}