Parallelizing sparse Matrix Solve for SPICE circuit simulation using FPGAs

Fine-grained dataflow processing of sparse matrix-solve computation (Ax¿ = b¿) in the SPICE circuit simulator can provide an order of magnitude performance improvement on modern FPGAs. Matrix solve is the dominant component of the simulator especially for large circuits and is invoked repeatedly dur...

Full description

Saved in:
Bibliographic Details
Main Authors: Kapre, N., DeHon, A.
Format: Conference Proceeding
Language:English
Subjects:
Circuit simulation
Computational modeling
Computer architecture
Concurrent computing
Field programmable gate arrays
Integrated circuit interconnections
Nonlinear equations
Runtime
Sparse matrices
SPICE
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fine-grained dataflow processing of sparse matrix-solve computation (Ax¿ = b¿) in the SPICE circuit simulator can provide an order of magnitude performance improvement on modern FPGAs. Matrix solve is the dominant component of the simulator especially for large circuits and is invoked repeatedly during the simulation, once for every iteration. We process sparse-matrix computation generated from the SPICE-oriented KLU solver in dataflow fashion across multiple spatial floating-point operators coupled to high-bandwidth on-chip memories and interconnected by a low-latency network. Using this approach, we are able to show speedups of 1.2-64× (geometric mean of 8.8×) for a range of circuits and benchmark matrices when comparing double-precision implementations on a 250 MHz Xilinx Virtex-5 FPGA (65 nm) and an Intel Core i7 965 processor (45 nm).
DOI:10.1109/FPT.2009.5377665