High-Performance Designs for Linear Algebra Operations on Reconfigurable Hardware
Authors:
- 24 June 2008
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 57 (8) , 1057-1071
Abstract
Numerical linear algebra operations are key primitives in scientific computing. Performance optimizations of such operations have been extensively investigated. With the rapid advances in technology, hardware acceleration of linear algebra applications using FPGAs (field programmable gate arrays) has become feasible. In this paper, we propose FPGA-based designs for several basic linear algebra operations, including dot product, matrix-vector multiplication, matrix multiplication and matrix factorization. By identifying the parameters for each operation, we analyze the trade-offs and propose a high-performance design. In the implementations of the designs, the values of the parameters are determined according to the hardware constraints, such as the available chip area, the size of available memory, the memory bandwidth, and the number of I/O pins. The proposed designs are implemented on Xilinx Virtex-II Pro FPGAs. Experimental results show that our designs scale with the available hardware resources. Also, the performance of our designs compares favorably with that of general-purpose processor based designs. We also show that with faster floating-point units and larger devices, the performance of our designs increases accordingly.Keywords
This publication has 31 references indexed in Scilit:
- Reconfigurable computing with multiscale data fusion for remote sensing2006
- 64-bit floating-point FPGA matrix multiplication2005
- Floating-point sparse matrix-vector multiply for FPGAs2005
- Automatic Tuning of PDGEMM Towards Optimal Performance2005
- Time and Energy Efficient Matrix Factorization Using FPGAs2003
- A fast parallel reed-solomon decoder on a reconfigurable architecture2003
- Algorithm Engineering for Parallel Computation2002
- Reconfigurable computingACM Computing Surveys, 2002
- Automated empirical optimizations of software and the ATLAS projectParallel Computing, 2001
- Basic Linear Algebra Subprograms for Fortran UsageACM Transactions on Mathematical Software, 1979