1DPPEQU(1)                LAPACK routine (version 3.2)                DPPEQU(1)
2
3
4

NAME

6       DPPEQU  -  computes  row  and column scalings intended to equilibrate a
7       symmetric positive definite matrix A in packed storage and  reduce  its
8       condition number (with respect to the two-norm)
9

SYNOPSIS

11       SUBROUTINE DPPEQU( UPLO, N, AP, S, SCOND, AMAX, INFO )
12
13           CHARACTER      UPLO
14
15           INTEGER        INFO, N
16
17           DOUBLE         PRECISION AMAX, SCOND
18
19           DOUBLE         PRECISION AP( * ), S( * )
20

PURPOSE

22       DPPEQU  computes row and column scalings intended to equilibrate a sym‐
23       metric positive definite matrix A in packed storage and reduce its con‐
24       dition  number  (with  respect  to the two-norm).  S contains the scale
25       factors, S(i)=1/sqrt(A(i,i)), chosen so that the scaled matrix  B  with
26       elements B(i,j)=S(i)*A(i,j)*S(j) has ones on the diagonal.  This choice
27       of S puts the condition number of B within a factor N of  the  smallest
28       possible condition number over all possible diagonal scalings.
29

ARGUMENTS

31       UPLO    (input) CHARACTER*1
32               = 'U':  Upper triangle of A is stored;
33               = 'L':  Lower triangle of A is stored.
34
35       N       (input) INTEGER
36               The order of the matrix A.  N >= 0.
37
38       AP      (input) DOUBLE PRECISION array, dimension (N*(N+1)/2)
39               The  upper  or lower triangle of the symmetric matrix A, packed
40               columnwise in a linear array.  The j-th column of A  is  stored
41               in  the array AP as follows: if UPLO = 'U', AP(i + (j-1)*j/2) =
42               A(i,j) for 1<=i<=j; if UPLO = 'L',  AP(i  +  (j-1)*(2n-j)/2)  =
43               A(i,j) for j<=i<=n.
44
45       S       (output) DOUBLE PRECISION array, dimension (N)
46               If INFO = 0, S contains the scale factors for A.
47
48       SCOND   (output) DOUBLE PRECISION
49               If  INFO  = 0, S contains the ratio of the smallest S(i) to the
50               largest S(i).  If SCOND >= 0.1 and AMAX is  neither  too  large
51               nor too small, it is not worth scaling by S.
52
53       AMAX    (output) DOUBLE PRECISION
54               Absolute  value  of  largest  matrix  element.  If AMAX is very
55               close to overflow or very close to underflow, the matrix should
56               be scaled.
57
58       INFO    (output) INTEGER
59               = 0:  successful exit
60               < 0:  if INFO = -i, the i-th argument had an illegal value
61               > 0:  if INFO = i, the i-th diagonal element is nonpositive.
62
63
64
65 LAPACK routine (version 3.2)    November 2008                       DPPEQU(1)
Impressum