1% SSMULT: sparse matrix multiplication (sparse times sparse) 2% 3% SSMULT computes C=A*B where A and B are sparse. It is typically faster 4% than C=A*B in MATLAB 7.4 (or earlier), and always uses less memory. 5% 6% ssmult - multiplies two sparse matrices. 7% ssmultsym - computes nnz(C), memory, and flops to compute C=A*B; A and B sparse. 8% 9% Compiling and testing: 10% ssmult_install - compiles, installs, and optionally tests ssmult. 11% ssmult_test - lengthy test of SSMULT and SSMULTSYM 12% sstest - exhaustive performance test for SSMULT. 13% sstest2 - exhaustive performance test for SSMULT. Requires ssget. 14% 15% Example: 16% C = ssmult(A,B) ; % computes C = A*B 17% 18% Copyright 2011, Timothy A. Davis, http://www.suitesparse.com 19