Subscribe
Issue No.12  December (2009 vol.58)
pp: 1
Jung Sub Kim , Dept. of Comput. Sci. & Eng., Pennsylvania State Univ., University Park, PA, USA
Lanping Deng , Dept. of Electr. Eng., Arizona State Univ., Tempe, AZ, USA
P. Mangalagiri , Dept. of Comput. Sci. & Eng., Pennsylvania State Univ., University Park, PA, USA
K. Irick , Dept. of Comput. Sci. & Eng., Pennsylvania State Univ., University Park, PA, USA
K. Sobti , Dept. of Electr. Eng., Arizona State Univ., Tempe, AZ, USA
M. Kandemir , Dept. of Comput. Sci. & Eng., Pennsylvania State Univ., University Park, PA, USA
V. Narayanan , Dept. of Comput. Sci. & Eng., Pennsylvania State Univ., University Park, PA, USA
C. Chakrabarti , Dept. of Electr. Eng., Arizona State Univ., Tempe, AZ, USA
N. Pitsianis , Dept. of Comput. Sci., Duke Univ., Durham, NC, USA
Xiaobai Sun , Dept. of Comput. Sci., Duke Univ., Durham, NC, USA
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TC.2009.78
ABSTRACT
This paper describes TANOR, an automated framework for designing hardware accelerators for numerical computation on reconfigurable platforms. Applications utilizing numerical algorithms on largesize data sets require highthroughput computation platforms. The focus is on Nbody interaction problems which have a wide range of applications spanning from astrophysics to molecular dynamics. The TANOR design flow starts with a MATLAB description of a particular interaction function, its parameters, and certain architectural constraints specified through a graphical user interface. Subsequently, TANOR automatically generates a configuration bitstream for a target FPGA along with associated drivers and control software necessary to direct the application from a host PC. Architectural exploration is facilitated through support for fully custom fixedpoint and floatingpoint representations in addition to standard number representations such as singleprecision floating point. Moreover, TANOR enables joint exploration of algorithmic and architectural variations in realizing efficient hardware accelerators. TANOR's capabilities have been demonstrated for three different Nbody interaction applications: the calculation of gravitational potential in astrophysics, the diffusion or convolution with Gaussian kernel common in image processing applications, and the force calculation with vectorvalued kernel function in molecular dynamics simulation. Experimental results show that TANORgenerated hardware accelerators achieve lower resource utilization without compromising numerical accuracy, in comparison to other existing custom accelerators.
INDEX TERMS
Hardware, Algorithm design and analysis, MATLAB, Field programmable gate arrays, Accuracy, Numerical analysis,numerical algorithms., Algorithms implemented in hardware, reconfigurable hardware, signal processing systems
CITATION
Jung Sub Kim, Lanping Deng, P. Mangalagiri, K. Irick, K. Sobti, M. Kandemir, V. Narayanan, C. Chakrabarti, N. Pitsianis, Xiaobai Sun, "An Automated Framework for Accelerating Numerical Algorithms on Reconfigurable Platforms Using Algorithmic/Architectural Optimization", IEEE Transactions on Computers, vol.58, no. 12, pp. 1, December 2009, doi:10.1109/TC.2009.78
REFERENCES
