Sparse-matrix solution is a dominant part of execution time in simulating VLSI circuits by a detailed simulation program such as SPICE. The paper develops a parallel-block partitionable sparse-matrix-solution algorithm which exploits sparsity at the matrix block level as well as within a nonzero block. An efficient mapping scheme to assign different matrix blocks to processors is developed which maximises concurrency and minimises communication between processors. Associated reordering and efficient sparse storage schemes are also developed. Implementation of this parallel algorithm is carried out on a transputer processor array which plugs into a PC bus. The sparse matrix solver is tested on matrices generated from a transistor-level expansion of ISCAS-85 benchmark logic circuits. Good acceleration is obtained for all benchmark matrices up to the number of transputers available.
References
-
-
1)
-
J.K. White ,
A.S. Vincentelli
.
(1987)
Relaxation techniques for the simulation of VLSI circuits.
-
2)
-
P. Cox ,
R. Burch
.
Direct circuit simulation algorithms for parallel processing.
IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.
,
6 ,
714 -
725
-
3)
-
V. Kumar ,
A. Grama ,
A. Gupta ,
G. Karypis
.
(1994)
Introduction to parallel computing, design and analysis of algorithms.
-
4)
-
Chilakpati, U.: `Parallel SOLVE for SPICE on a network of workstations', MS project report, 1995.
-
5)
-
Trotter, J.A., Agrawal, P.: `Circuit simulation algorithms on a distributed memory multiprocessorsystem', Proceedings of ICCAD-90, 1990, p. 438–441.
-
6)
-
Sangiovanni-Vincentelli, A.: `A new tearing approach-node tearing nodal analysis', Proceedings of the IEEE international symposium on Circuitsand systems, 1977, p. 143–145.
-
7)
-
H.M. Markowitz
.
The elimination form of the inverse and its application to linear programming.
Manage. Sci.
,
255 -
269
-
8)
-
A. Mahmood ,
S. Sparks ,
W.I. Baker
.
Systolic algorithms for solving a sparse system of linear equations incircuit simulation.
Integr. VLSI J.
,
1 ,
83 -
107
-
9)
-
B. Greer
.
Converting SPICE to vector code.
VLSI Syst. Des.
,
1
-
10)
-
Vladimirescu, A., Pederson, D.O.: `Circuit simulation on vector processors', Proceedings of the IEEE ICCC, 1982, p. 172–175.
-
11)
-
J.J. Navarro
.
Partitioning: an essential step in mapping algorithms into systolic arrayprocessors.
Computer
,
77 -
89
-
12)
-
Trotter, J.A., Agrawal, P.: `Fast overlapped scattered array storage schemes for sparse matrices', Proceedings of the ICCAD, 1990, p. 450–453.
-
13)
-
Sadayappan, P.: `Efficient sparse matrix factorization for circuit simulation on vectorsupercomputers', Proceedings of the 26th Design automation conference, 1989, p. 13–18.
-
14)
-
Nagel, L.W.: `SPICE2: a computer program to simulate semiconductor circuits', ERL-M520, memorandum, May 1975.
-
15)
-
Chu, Y.: `Parallel solution of sparse matrix equations in SPICE on a transputerarray', MS project report, 1995.
-
16)
-
K. Hwang ,
Y.H. Cheng
.
Partitioned matrix algorithms for VLSI arithmetic systems.
IEEE Trans.
,
12 ,
1215 -
1224
-
17)
-
Eisenstat, S.C.: `Yale sparse matrix package II: the nonsymmetric codes', 114, Research report, 1977.
-
18)
-
P. Sadayappan ,
V. Visvanathan
.
Circuit simulation on shared memory multiprocessors.
IEEE Trans.
,
12 ,
1634 -
1642
-
19)
-
Yang, G.C.: `PARASPICE: a parallel circuit simulator for shared-memory multiprocessors', Proceedings of the 27th Design automation conference, 1990, p. 400–405.
http://iet.metastore.ingenta.com/content/journals/10.1049/ip-cds_19971566
Related content
content/journals/10.1049/ip-cds_19971566
pub_keyword,iet_inspecKeyword,pub_concept
6
6