Title :
Power performance analysis of 3-D finite element mesh refinement with tetrahedra by CUDA/MPI on multi-core and GPU architecture
Author :
Ren, Da Qi ; Giannacopoulos, Dennis D. ; Suda, Reiji
Author_Institution :
Dept. of Comput. Sci., Univ. of Tokyo, Tokyo, Japan
Abstract :
Software power performance tuning handles the critical design constraints at the level of software that run beyond hardware platforms composed of large numbers of power-hungry components. The power dissipation of a Single Instruction Multiple Data (SIMD) program such as FEM mesh refinement is highly dependent on its underlying algorithm and the power-consuming features of hardware Processing Elements (PE). This contribution presents a practical methodology for modeling and analyzing the power performance of parallel 3-D FEM mesh refinement on CUDA/MPI architecture based on detailed software prototypes and power parameters in order to predict the power functionality and runtime behavior of the algorithm, optimize the program design and thus achieve the best power efficiency.
Keywords :
computer graphic equipment; coprocessors; finite element analysis; multiprocessing systems; parallel programming; power aware computing; CUDA/MPI architecture; GPU architecture; hardware processing elements; parallel 3D finite element mesh refinement; single instruction multiple data program; software power performance analysis; tetrahedra; Algorithm design and analysis; Computer architecture; Finite element methods; Hardware; Performance analysis; Power dissipation; Predictive models; Runtime; Software performance; Software prototyping;
Conference_Titel :
Electromagnetic Field Computation (CEFC), 2010 14th Biennial IEEE Conference on
Conference_Location :
Chicago, IL
Print_ISBN :
978-1-4244-7059-4
DOI :
10.1109/CEFC.2010.5481537
