Title :
Sparse Matrix Computations on Reconfigurable Hardware
Author :
Prasanna, Viktor K. ; Morris, Gerald R.
Author_Institution :
Univ. of Southern California, CA
fDate :
3/1/2007 12:00:00 AM
Abstract :
Using a high-level-language to hardware-description-language compiler and some novel architectures and algorithms to map two well-known double-precision floating-point sparse matrix iterative-linear-equation solvers - the Jacobi and conjugate gradient methods - onto a reconfigurable computer achieves more than a twofold speedup over software
Keywords :
conjugate gradient methods; hardware description languages; high level languages; program compilers; reconfigurable architectures; sparse matrices; Jacobi method; conjugate gradient method; double-precision floating-point sparse matrix iterative-linear-equation solver; hardware-description-language compiler; high-level-language; reconfigurable computer hardware; Algorithm design and analysis; Binary trees; Character generation; Clocks; Degradation; Field programmable gate arrays; Hardware design languages; Pipelines; Runtime; Sparse matrices; Conjugate gradient method; FPGAs; Jacobi method; reconfigurable computing;
DOI :
10.1109/MC.2007.103