261 
v add(V,ll,V); 
sp_vm_mlt(P,V,u); 
mO=in_prod(u,V); 
mO=sqrt(mO/(A->m-A->n));//mO out 
spfree(AT); spfree(N); spiree(Temp); 
} 
By experiment, when the order number of matrix A is about ten 
thousands, the programme is processed well. 
REFERENCE 
[1] Numerical Analysis. Rainer Kress. Springer-Verlag. 1991 
[2] I.S.Duff, A.M.Erisman, and J.K.Reid. Direct Methods for 
Sparse Matrices. London: Oxford Univ. Press, 1986. 
[3] J.W.H.Liu. The Multifrontal Method for Sparse Matrix 
Solution: Theory and Practice. SIAM Rev., 34 (1992), pp. 82— 
109. 
[4] T.A.Davis.A column pre-ordering strategy for the 
unsymmetric-pattem multifrontal method, ACM Trans. Math. 
Software, vol 30, no. 2, pp. 165-195, 2004. 
[5] N.J.Higham.Accuracy and Stability of Numerical 
Algorithms. SIAM,2002 
[6] G..H.Golob, C.F.Van loan. Matrix Computations. The Johns 
Hopkins University Press. 1996 
[7] J.W.H.Liu. The Multifrontal Method for Sparse Matrix 
Solution: Theory and Practice. SIAM Rev., 34 (1992), pp. 82— 
109. 
[8] Fast PageRank Computation via a Sparse Linear System 
(Extended Abstract) Gianna M. Del Corso,Antonio Gulli, 
Francesco Romani. 
[9] Y.Saad, Iterative Methods for Sparse Linear Systems, PWS, 
Boston, 1996 
[10] http://www.cs.ut.ee/~toomas_l/linalg/lin2/node5.html 
[1 l]http://www.cs.ut.ee/~toomas_l/linalg/lin2/nodel .html 
[ 12]http://www.math.purdue.edu 
[13] http://www.cs.ut.ee/~toomas_l/linalg/lin2/nodel0.html#SE 
CTIONOOO12500000000000000 
[14] http://www.cs.ut.ee/~toomas_l/linalg/lin2/nodel 1 .html 
[ 15]http://web.mit.edu/be.400/www/SVD/Singular_Value_Dec 
omposition.htm Singular Value Decomposition (SVD) tutorial 
[ 16]http://www.xs4all.nl/~hkuiper/cwmtx/cwmtx.html 
[17]http://linux.about.com/cs/linuxl01/g/meschach.htm 
[ 18]http://www-128. ibm.com/developerworks/linux/library/1- 
matrix.html 
[ 19]http://www.mathworks.com/company/newsletters/news_no 
tes/clevescomer/oct02_cleve.html 
[20]http://www.intel.com/cd/00/00/21/93/219302_lin_relnotes. 
pdf Intel® Math Kernel Library 9.0 for Linux* Release Notes 
[21 ]http://www.math.ethz.ch/isg/computing/sophokles/pardiso