High-Performance Architecture for the Conjugate Gradient Solver on FPGAs
- 29 August 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems II: Express Briefs
- Vol. 60 (11), 791-795
- https://doi.org/10.1109/tcsii.2013.2278111
Abstract
The conjugate gradient (CG) solver is an important algorithm for solving the symmetric positive define systems. However, existing CG architectures on field-programmable gate arrays (FPGAs) either need aggressive zero padding or can only be applied for small matrices and particular matrix sparsity patterns. This brief proposes a high-performance architecture for the CG solver on FPGAs, which can handle sparse linear systems with arbitrary size and sparsity pattern. Furthermore, it does not need aggressive zero padding. Our CG architecture mainly consists of a high-throughput sparse matrix-vector multiplication design including a multi-output adder tree, a reduction circuit, and a sum sequencer. Our experimental results demonstrate that our CG architecture can achieve speedup of 4.62X-9.24X on a Virtex5-330 FPGA, relative to a software implementation.Keywords
This publication has 11 references indexed in Scilit:
- Towards a Universal FPGA Matrix-Vector Multiplication ArchitecturePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2012
- The university of Florida sparse matrix collectionACM Transactions on Mathematical Software, 2011
- An I/O Bandwidth-Sensitive Sparse Matrix-Vector Multiplication Engine on FPGAsIEEE Transactions on Circuits and Systems I: Regular Papers, 2011
- Sparse Matrix-Vector Multiplication on a Reconfigurable Supercomputer with ApplicationACM Transactions on Reconfigurable Technology and Systems, 2010
- Exploiting Matrix Symmetry to Improve FPGA-Accelerated Conjugate GradientPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2009
- Sparse Matrix-Vector Multiplication Design on FPGAsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2007
- Pipelined Mixed Precision Algorithms on FPGAs for Fast and Accurate PDE Solvers from Low Precision ComponentsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2006
- FPGA Implementation of the Conjugate Gradient MethodLecture Notes in Computer Science, 2006
- Sparse Matrix-Vector multiplication on FPGAsPublished by Association for Computing Machinery (ACM) ,2005
- A fast radix-4 division algorithm and its architectureInternational Conference on Acoustics, Speech, and Signal Processing (ICASSP), 1995