FPGA Implementation of the Conjugate Gradient Method
The rational fraction number system is proposed to solve the algebraic problems in FPGA devices. The fraction number consists of the n-bit integer numerator and the n -bit integer denominator, and can represent numbers with 2n bit mantissa. Experimental linear equation system solver was developed in FPGA device, which implements the recursive conjugate gradient method. Its hardware arithmetic unit can calculate addition, multiplication, and division of fraction numbers with n=35 in a pipelined mode. The proposed unit operates with the band matrices with the dimensions up to 3500.
Unable to display preview. Download preview PDF.
- 1.Underwood, K.D., Hemmert, K.S.: Closing the Gap: CPU and FPGA Trends in sustained Floating Point BLAS Performance. In: Proc. IEEE Symp. Field Programmable Custom Computing Machines, FCCM-2004 (2004)Google Scholar
- 2.Dou, Y., Vassiliadis, S., Kuzmanov, G.K., Gaydadjiev, G.N.: 64-bit Floating point FPGA Matrix Multiplication. In: ACM/SIGDA 13-th Int. Symp. on Field Programmable Gate Arrays, FPGA-2005, February 2005, pp. 86–95 (2005)Google Scholar
- 4.Golub, G.G., Van Loan, C.F.: Matrix Computations, 2nd edn., p. 642. J. Hopkins Univ. Press (1989)Google Scholar
- 6.Hintchin, A.Y.: Chained Fractions (in Russian).: 3d edn., p. 112. Nauka, Moscow (1978)Google Scholar
- 7.Irvin, M.J., Smith, D.R.: A rational arithmetic processor. In: Proc. 5-th Symp. Comput. Arithmetic, pp. 241–244 (1981)Google Scholar
- 8.Holmes, W.N.: Composite Arithmetic: Proposal for a new Standard. Computer N3, 65–72 (March 1997)Google Scholar