We restrict ourselves to iterative methods, which work by repeatedly improving an approximate solution until it is accurate enough. These methods access the coefficient matrix A of the linear system only via the matrix-vector product y = A · x (and perhaps z = AT · x). Thus the user need only supply a subroutine for computing y (and perhaps z) given x, which permits full exploitation of the sparsity or other special structure of A.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|