ascend4/solver/mtx_reorder.h

/*
 *  mtx: Ascend Sparse Matrix Package
 *  by Benjamin Andrew Allan
 *  Derived from mtx by Karl Michael Westerberg
 *  Created: 5/3/90
 *  Version: $Revision: 1.4 $
 *  Version control file: $RCSfile: mtx_reorder.h,v $
 *  Date last modified: $Date: 1997/07/18 12:15:17 $
 *  Last modified by: $Author: mthomas $
 *
 *  This file is part of the SLV solver.
 *
 *  Copyright (C) 1996 Benjamin Andrew Allan
 *
 *  The SLV solver is free software; you can redistribute
 *  it and/or modify it under the terms of the GNU General Public License as
 *  published by the Free Software Foundation; either version 2 of the
 *  License, or (at your option) any later version.
 *
 *  The SLV solver is distributed in hope that it will be
 *  useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 *  General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License along with
 *  the program; if not, write to the Free Software Foundation, Inc., 675
 *  Mass Ave, Cambridge, MA 02139 USA.  Check the file named COPYING.
 *  COPYING is found in ../compiler.
 */
#ifndef __MTX_REORDER_H_SEEN__
#define __MTX_REORDER_H_SEEN__
/* requires #include "mtx.h" */

enum mtx_reorder_method {
  mtx_UNKNOWN,  /* junk method */
  mtx_SPK1, /* Stadtherr's SPK1 reordering */
  mtx_TSPK1,    /* transpose of Stadtherr's SPK1 reordering */
  mtx_NATURAL   /* kinda pointless, don't you think? */
};

extern int mtx_reorder(mtx_matrix_t,mtx_region_t *, enum mtx_reorder_method);
/**
 ***  mtx_reorder(sys,region,rmeth)
 ***  mtx_matrix_t mtx;
 ***  mtx_region_t *region;
 ***  enum mtx_reorder_method;
 ***
 ***  The specified region of the coefficient matrix is reordered.
 ***  The specified region
 ***  is assumed to contain only nonempty rows and columns and have a full
 ***  diagonal.
 ***  If the matrix in is from a nonlinear system, the
 ***  pattern in the matrix should probably be the structural one (as
 ***  opposed to the numerically derived incidence which may be less.)
 ***
 ***  If you use the numerically derived incidence, you will need to reorder
 ***  before every factorization. This is generally not cost effective.
 ***  If region given is mtx_ENTIRE_MATRIX, a search will be done to find
 ***  an appropriate bounding region in the coefficient mtx. This is
 ***  not a particularly cheap search.
 *** 
 ***  The reorder call should be done (once) at
 ***  the beginning of any series of linear solutions being performed on
 ***  on structurally identical matrices.
 ***
 ***  We HATE mtx_ENTIRE_MATRIX as the input region.
 ***
 ***  Brief notes on the reorderings available.
 ***    SPK1:  
 ***   TSPK1: Both these operate only on a square region which is of. 
 ***          full rank (in particular it should have a 0 free diagonal).
 ***          If you give us a bogus region, we will Try to make something
 ***          decent of it, but good results are improbable.
 ***   Natural:  Blesses the system and does nothing.
 ***             Again, the rows/cols not in the diagonal are dependent.
 ***
 ***  On reordering in general: 'Optimal' reordering is an NP complete
 ***  task. Real reordering methods are heuristic and tend to break down
 ***  when the matrix being reordered is in some sense 'large' so that
 ***  the tie-breaking and other heuristics fail.
 ***
 ***  Return 0 if ok, 1 if bad input detected, 2 if unable to do.
 **/
#endif /* __MTX_REORDER_H_SEEN__ */
1	aw0a	1	/*
2			* mtx: Ascend Sparse Matrix Package
3			* by Benjamin Andrew Allan
4			* Derived from mtx by Karl Michael Westerberg
5			* Created: 5/3/90
6			* Version: $Revision: 1.4 $
7			* Version control file: $RCSfile: mtx_reorder.h,v $
8			* Date last modified: $Date: 1997/07/18 12:15:17 $
9			* Last modified by: $Author: mthomas $
10			*
11			* This file is part of the SLV solver.
12			*
13			* Copyright (C) 1996 Benjamin Andrew Allan
14			*
15			* The SLV solver is free software; you can redistribute
16			* it and/or modify it under the terms of the GNU General Public License as
17			* published by the Free Software Foundation; either version 2 of the
18			* License, or (at your option) any later version.
19			*
20			* The SLV solver is distributed in hope that it will be
21			* useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
22			* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
23			* General Public License for more details.
24			*
25			* You should have received a copy of the GNU General Public License along with
26			* the program; if not, write to the Free Software Foundation, Inc., 675
27			* Mass Ave, Cambridge, MA 02139 USA. Check the file named COPYING.
28			* COPYING is found in ../compiler.
29			*/
30			#ifndef __MTX_REORDER_H_SEEN__
31			#define __MTX_REORDER_H_SEEN__
32			/* requires #include "mtx.h" */
33
34			enum mtx_reorder_method {
35			mtx_UNKNOWN, /* junk method */
36			mtx_SPK1, /* Stadtherr's SPK1 reordering */
37			mtx_TSPK1, /* transpose of Stadtherr's SPK1 reordering */
38			mtx_NATURAL /* kinda pointless, don't you think? */
39			};
40
41			extern int mtx_reorder(mtx_matrix_t,mtx_region_t *, enum mtx_reorder_method);
42			/**
43			*** mtx_reorder(sys,region,rmeth)
44			*** mtx_matrix_t mtx;
45			*** mtx_region_t *region;
46			*** enum mtx_reorder_method;
47			***
48			*** The specified region of the coefficient matrix is reordered.
49			*** The specified region
50			*** is assumed to contain only nonempty rows and columns and have a full
51			*** diagonal.
52			*** If the matrix in is from a nonlinear system, the
53			*** pattern in the matrix should probably be the structural one (as
54			*** opposed to the numerically derived incidence which may be less.)
55			***
56			*** If you use the numerically derived incidence, you will need to reorder
57			*** before every factorization. This is generally not cost effective.
58			*** If region given is mtx_ENTIRE_MATRIX, a search will be done to find
59			*** an appropriate bounding region in the coefficient mtx. This is
60			*** not a particularly cheap search.
61			***
62			*** The reorder call should be done (once) at
63			*** the beginning of any series of linear solutions being performed on
64			*** on structurally identical matrices.
65			***
66			*** We HATE mtx_ENTIRE_MATRIX as the input region.
67			***
68			*** Brief notes on the reorderings available.
69			*** SPK1:
70			*** TSPK1: Both these operate only on a square region which is of.
71			*** full rank (in particular it should have a 0 free diagonal).
72			*** If you give us a bogus region, we will Try to make something
73			*** decent of it, but good results are improbable.
74			*** Natural: Blesses the system and does nothing.
75			*** Again, the rows/cols not in the diagonal are dependent.
76			***
77			*** On reordering in general: 'Optimal' reordering is an NP complete
78			*** task. Real reordering methods are heuristic and tend to break down
79			*** when the matrix being reordered is in some sense 'large' so that
80			*** the tie-breaking and other heuristics fail.
81			***
82			*** Return 0 if ok, 1 if bad input detected, 2 if unable to do.
83			**/
84			#endif /* __MTX_REORDER_H_SEEN__ */