International Journal of Computer Applications

© 2011 by IJCA Journal

Volume 35 - Number 8

Year of Publication: 2011

10.5120/4419-6140 |

D N Sonawane and M S Sutaone. Article: High Throughput Iterative VLSI Architecture for Cholesky Factorization Based Matrix Inversion. *International Journal of Computer Applications* 35(8):10-15, December 2011. Full text available. BibTeX

@article{key:article, author = {D. N. Sonawane and M. S. Sutaone}, title = {Article: High Throughput Iterative VLSI Architecture for Cholesky Factorization Based Matrix Inversion}, journal = {International Journal of Computer Applications}, year = {2011}, volume = {35}, number = {8}, pages = {10-15}, month = {December}, note = {Full text available} }

Cholesky factorization is the computationally most expensive step in numerically solving positive definite systems. Due to inherently recursive computation process and associated floating point division and square root operations in Cholesky factorization, it is very difficult to obtain acceleration by exploiting parallelism on FPGA’s. To solve this problem, approach suggests iterative architecture with parallelly fetching the matrix elements using customized Diagonal Processing Elements (DPU), Non Diagonal Processing Elements (NDPU) and Triangular Processing Elements (TPU) as computational processing units. The use of LNS approach using LUT technique for floating point square root and division arithmetic eventually improves resource and clock cycle utilization. Scheme is implemented using Xilinx Virtex-4 FPGA and achieves 0.032µs clock latency and obtained a throughput of 31.25Mupdates/s operating at 125 MHz for 4x4 matrix inversion problem.

- Antonio Roldao and George A. Constantinides, “A High Throughput FPGA-Based Floating Point Conjugate Gradient Implementation for Dense Matrices”, ACM Transaction on Reconfigurable Technoogy Systems, Vol. 3 (1), Jan. 2010.
- J. Sun, et.al, “High Performance Mixed-Precision Linear Solver for FPGAs,” IEEE Transaction on Computers, vol. 57, (12) Dec. 2008
- D. Yang, et. al, "Performance Comparison of Cholesky Decomposition on GPUs and FPGAs," Symposium on Application Accelerators in High Performance Computing (SAAHPC), knoxville, TN, 2010.
- D. Yang, G. D. Peterson, and H. Li, "High Performance Reconfigurable Computing for Cholesky Decomposition," Symposium on Application Accelerators in High Performance Computing (SAAHPC), UIUC, July, 2009.
- D. Yang, et.al, "An FPGA Implementation for Solving Least Square Problem," The 17th IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM), Napa, California, April. 2009.
- Antonio Roldao Lopes and George A. Constantinides, “A High Throughput FPGA-Based Floating Point Conjugate GradientImplementation”, Proceedings of the 4th international workshop onReconfigurable Computing: Architectures, Tools and Applications, Springer-Verlag, Berlin, Heidelberg,2008, pp 75-86
- O. Maslennikow, et.al, “Parallel Implementation of Cholesky LLT-Algorithm in FPGA-Based Processor,” Proceedings of the 7th international conference on Parallel processing and applied mathematics Springer- Verlag Berlin Heidelberg, , June-2007pp. 137-147
- Maslennikow,et.al.,"Implementation of Cholesky LLT-Decomposition Algorithm in FPGA-Based Rational Fraction Parallel Processor," Mixed Design of Integrated Circuits and Systems, 14th International conference on mixed design (MIXDES 2007),Ciechocinek, POLAND, 21-23 June 2007, pp.287-292
- Burian, A. et.al , "A fixed-point implementation of matrix inversion using Cholesky decomposition," Micro-NanoMechatronics and Human Science, Proceedings of the 46th IEEE International Midwest Symposium on In Circuits and Systems, 2003. MWSCAS , Dec. 2003, pp. 1431- 1434
- S. Haridas, “FPGA Implementation of a Cholesky Algorithm for a Shared-Memory Multiprocessor Architecture,”, A masters thesis, submitted to the faculty of New Jersey Institute of Technology, May 2003.
- Garcia et al. “LNS Architectures for Embedded Model PredictiveControl Processors”In Proceedings of International Conference onCompilers, Architecture, and Synthesis for Embedded Systems,CASES 2004, Washington DC, USA, Sep. 2004
- A book on “VLSI Digital Signal Processing System Design and Implementation”, by K. K. Parhi, Wiley, 1999.