Subversion Repositories slepc-dev

Rev

Details | Compare with Previous | Last modification | View Log | RSS feed

Rev Author Line No. Line
6 dsic.upv.es!jroman 1
 
2
 
3
        SLEPc: Scalable Library for Eigenvalue Problem Computations
4
        ===========================================================
5
 
6
                            Vicente Hernandez
7
                              Jose E. Roman
237 dsic.upv.es!jroman 8
                              Andres Tomas
6 dsic.upv.es!jroman 9
                              Vicent Vidal
10
 
11
        GRyCAP - Grupo de Redes y Computacion de Altas Prestaciones
12
                Universidad Politecnica de Valencia, Spain
13
 
14
 
15
 Overview
16
 --------
17
 
18
 SLEPc, the Scalable Library for Eigenvalue Problem Computations, is a software
19
 package for  the  solution of large sparse  eigenvalue  problems  on  parallel
20
 computers. It  can be used for the solution of problems formulated  in  either
21
 standard  or  generalized form, as well as other related problems such as  the
22
 singular value decomposition.
23
 
24
 The  emphasis  of the software is on methods and  techniques  appropriate  for
25
 problems  in  which  the  associated matrices are sparse, for  example,  those
26
 arising after the discretization of partial differential equations. Therefore,
27
 most  of  the methods offered by the library are projection methods  or  other
28
 methods  with similar properties. Some of these methods are  Arnoldi,  Lanczos
29
 and Subspace Iteration, to name a few. SLEPc implements these basic methods as
30
 well  as more sophisticated algorithms. It also provides built-in support  for
31
 spectral transformations such as shift-and-invert.
32
 
33
 SLEPc is a general library in the sense that it covers standard and genralized
34
 eigenvalue  problems,  both Hermitian and non-Hermitian, with either  real  or
35
 complex arithmetic.
36
 
37
 SLEPc is built on top of PETSc, the Portable Extensible Toolkit for Scientific
38
 Computation (http://www.mcs.anl.gov/petsc). It can be considered an  extension
39
 of  PETSc  providing  all the functionality  necessary  for  the  solution  of
40
 eigenvalue  problems.  This means that PETSc must be previously  installed  in
41
 order  to  use SLEPc. PETSc users will find SLEPc very easy to use,  since  it
42
 enforces the same programming paradigm. For those users which are not familiar
43
 with  PETSc yet, our recommendation is to fully understand its basic  concepts
44
 before proceeding with SLEPc.
45
 
46
 
47
 Documentation
48
 -------------
49
 
50
 The  Users Manual is included in the SLEPc distribution file. It can be  found
51
 in directory  'docs' and it contains a general description of the capabilities
52
 of  the software.  The manual does not include detailed reference  information
53
 about  individual SLEPc routines. This information is provided in the form  of
54
 man pages in HTML format (see 'docs/manualpages').
55
 
56
 
57
 Installation
58
 ------------
59
 
60
 The installation procedure of SLEPc is very similar to that of PETSc. Briefly,
61
 the  environment variable $SLEPC_DIR must be set and then  the  libraries  are
62
 built with the command 'make BOPT=O', for example. For this to work correctly,
63
 variables $PETSC_DIR and $PETSC_ARCH must also be set appropriately.
64
 
65
 More detailed information about installation can be found in the Users  Manual
66
 or in the SLEPc home page, including instructions for configuring SLEPc to use
67
 external libraries such as ARPACK.
68
 
69
 
70
 More Information
71
 ----------------
72
 
73
 Additional  information can be found in the SLEPc home page at  the  following
74
 address:
75
 
76
                  http://www.grycap.upv.es/slepc
77
 
78
 Among other things, this site includes information about:
79
 
80
  - How to contact the authors for support.
81
  - Download page including available patches (if any).
82
  - On-line documentation.
83
  - Mailing list for announcement of new releases.
84
 
85
 All questions/comments should be directed to slepc-maint@grycap.upv.es.
86
 
87
 
88
 Acknowledgements
89
 ----------------
90
 
91
 The development of SLEPc has been partially supported by the following grants:
92
 
93
 - Oficina de Ciencia y Tecnologia, Generalitat Valenciana, CTIDB/2002/54.
94
 
95
 
96
 Conditions of use
97
 -----------------
98
 
237 dsic.upv.es!jroman 99
 Copyright (c) 2002-2004, Universidad Politecnica de Valencia, Spain
6 dsic.upv.es!jroman 100
 
101
 This  software is provided 'as is', with absolutely no warranty, expressed  or
102
 implied.  Any use is at your own risk. In no event shall the authors be liable
103
 for  any direct or indirect damages arising in any way out of the use of  this
104
 software.
105
 
106
 The  user will acknowledge (using references [1] or [2]) the  contribution  of
107
 SLEPc  in any publication of material dependent upon the use of  the  package.
108
 The  user  will reasonably endeavour to notify the authors of the  package  of
109
 this publication.
110
 
111
 The  user can modify the code but, at no time shall the right or title to  all
112
 or  any  part  of  this  package pass to the user. Contributions  are  welcome
113
 relating  to any alteration or addition made to this package for the  purposes
114
 of  extending the capabilities or enhancing its performance.  Credit  will  be
115
 given in the documentation for contributed code.
116
 
117
 This  package (or a modified version) may not be sold. It is freely  available
118
 and  it can be redistributed provided that redistribution comprises all  files
119
 including this copyright notice. No license is required for research use.  For
120
 commercial use, written permission must be granted by the authors.
121
 
122
 
123
 [1] V. Hernandez, J. E. Roman and V. Vidal (2003),
124
     SLEPc: Scalable Library for Eigenvalue Problem Computations,
125
     Lecture Notes in Computer Science, 2565, pp. 377-391.
126
 
127
 [2] V. Hernandez, J. E. Roman and V. Vidal (2002),
128
     SLEPc Users Manual, Technical Report DSIC-II/24/02,
129
     Universidad Politecnica de Valencia, Spain.
130
 
131