DocumentCode :
598595
Title :
Early evaluation of directive-based GPU programming models for productive exascale computing
Author :
Seyong Lee ; Vetter, Jeffrey S.
Author_Institution :
Oak Ridge Nat. Lab., Oak Ridge, TN, USA
fYear :
2012
fDate :
10-16 Nov. 2012
Firstpage :
1
Lastpage :
11
Abstract :
Graphics Processing Unit (GPU)-based parallel computer architectures have shown increased popularity as a building block for high performance computing, and possibly for future Exascale computing. However, their programming complexity remains as a major hurdle for their widespread adoption. To provide better abstractions for programming GPU architectures, researchers and vendors have proposed several directive-based GPU programming models. These directive-based models provide different levels of abstraction, and required different levels of programming effort to port and optimize applications. Understanding these differences among these new models provides valuable insights on their applicability and performance potential. In this paper, we evaluate existing directive-based models by porting thirteen application kernels from various scientific domains to use CUDA GPUs, which, in turn, allows us to identify important issues in the functionality, scalability, tunability, and debuggability of the existing models. Our evaluation shows that directive-based models can achieve reasonable performance, compared to hand-written GPU codes.
Keywords :
graphics processing units; parallel architectures; parallel programming; software performance evaluation; CUDA GPUs; GPU architecture programming; abstraction levels; directive-based GPU programming model early evaluation; exascale computing; graphics processing unit-based parallel computer architectures; hand-written GPU codes; high performance computing; performance potential; productive exascale computing; program debuggability; program functionality; program scalability; program tunability; programming complexity; programming effort levels; Computational modeling; Computer architecture; Data models; Graphics processing units; Kernel; Optimization; Programming;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
High Performance Computing, Networking, Storage and Analysis (SC), 2012 International Conference for
Conference_Location :
Salt Lake City, UT
ISSN :
2167-4329
Print_ISBN :
978-1-4673-0805-2
Type :
conf
DOI :
10.1109/SC.2012.51
Filename :
6468490
Link To Document :
بازگشت