|
d@0
|
1 <html lang="en">
|
|
d@0
|
2 <head>
|
|
d@0
|
3 <title>FFTW MPI Wisdom - FFTW 3.2alpha3</title>
|
|
d@0
|
4 <meta http-equiv="Content-Type" content="text/html">
|
|
d@0
|
5 <meta name="description" content="FFTW 3.2alpha3">
|
|
d@0
|
6 <meta name="generator" content="makeinfo 4.8">
|
|
d@0
|
7 <link title="Top" rel="start" href="index.html#Top">
|
|
d@0
|
8 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
|
|
d@0
|
9 <link rel="prev" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes" title="FFTW MPI Transposes">
|
|
d@0
|
10 <link rel="next" href="Avoiding-MPI-Deadlocks.html#Avoiding-MPI-Deadlocks" title="Avoiding MPI Deadlocks">
|
|
d@0
|
11 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
|
|
d@0
|
12 <!--
|
|
d@0
|
13 This manual is for FFTW
|
|
d@0
|
14 (version 3.2alpha3, 14 August 2007).
|
|
d@0
|
15
|
|
d@0
|
16 Copyright (C) 2003 Matteo Frigo.
|
|
d@0
|
17
|
|
d@0
|
18 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
d@0
|
19
|
|
d@0
|
20 Permission is granted to make and distribute verbatim copies of
|
|
d@0
|
21 this manual provided the copyright notice and this permission
|
|
d@0
|
22 notice are preserved on all copies.
|
|
d@0
|
23
|
|
d@0
|
24 Permission is granted to copy and distribute modified versions of
|
|
d@0
|
25 this manual under the conditions for verbatim copying, provided
|
|
d@0
|
26 that the entire resulting derived work is distributed under the
|
|
d@0
|
27 terms of a permission notice identical to this one.
|
|
d@0
|
28
|
|
d@0
|
29 Permission is granted to copy and distribute translations of this
|
|
d@0
|
30 manual into another language, under the above conditions for
|
|
d@0
|
31 modified versions, except that this permission notice may be
|
|
d@0
|
32 stated in a translation approved by the Free Software Foundation.
|
|
d@0
|
33 -->
|
|
d@0
|
34 <meta http-equiv="Content-Style-Type" content="text/css">
|
|
d@0
|
35 <style type="text/css"><!--
|
|
d@0
|
36 pre.display { font-family:inherit }
|
|
d@0
|
37 pre.format { font-family:inherit }
|
|
d@0
|
38 pre.smalldisplay { font-family:inherit; font-size:smaller }
|
|
d@0
|
39 pre.smallformat { font-family:inherit; font-size:smaller }
|
|
d@0
|
40 pre.smallexample { font-size:smaller }
|
|
d@0
|
41 pre.smalllisp { font-size:smaller }
|
|
d@0
|
42 span.sc { font-variant:small-caps }
|
|
d@0
|
43 span.roman { font-family:serif; font-weight:normal; }
|
|
d@0
|
44 span.sansserif { font-family:sans-serif; font-weight:normal; }
|
|
d@0
|
45 --></style>
|
|
d@0
|
46 </head>
|
|
d@0
|
47 <body>
|
|
d@0
|
48 <div class="node">
|
|
d@0
|
49 <p>
|
|
d@0
|
50 <a name="FFTW-MPI-Wisdom"></a>
|
|
d@0
|
51 Next: <a rel="next" accesskey="n" href="Avoiding-MPI-Deadlocks.html#Avoiding-MPI-Deadlocks">Avoiding MPI Deadlocks</a>,
|
|
d@0
|
52 Previous: <a rel="previous" accesskey="p" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes">FFTW MPI Transposes</a>,
|
|
d@0
|
53 Up: <a rel="up" accesskey="u" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI">Distributed-memory FFTW with MPI</a>
|
|
d@0
|
54 <hr>
|
|
d@0
|
55 </div>
|
|
d@0
|
56
|
|
d@0
|
57 <h3 class="section">6.8 FFTW MPI Wisdom</h3>
|
|
d@0
|
58
|
|
d@0
|
59 <p><a name="index-wisdom-380"></a><a name="index-saving-plans-to-disk-381"></a>
|
|
d@0
|
60 FFTW's “wisdom” facility (see <a href="Words-of-Wisdom_002dSaving-Plans.html#Words-of-Wisdom_002dSaving-Plans">Words of Wisdom-Saving Plans</a>) can
|
|
d@0
|
61 be used to save MPI plans as well as to save uniprocessor plans.
|
|
d@0
|
62 However, for MPI there are several unavoidable complications.
|
|
d@0
|
63
|
|
d@0
|
64 <p>First, the MPI standard does not guarantee that every process can
|
|
d@0
|
65 perform file I/O (at least, not using C stdio routines)—only the
|
|
d@0
|
66 process of rank <code>MPI_IO</code> can in general (unless <code>MPI_IO ==
|
|
d@0
|
67 MPI_ANY_SOURCE</code>). (In practice, <code>MPI_IO</code> is commonly assumed to
|
|
d@0
|
68 be zero, i.e. at least process 0 can perform I/O, since otherwise a
|
|
d@0
|
69 single-process run could not perform I/O.) So, if we want to export
|
|
d@0
|
70 the wisdom from a single process to a file, we must first export the
|
|
d@0
|
71 wisdom to a string, then send it to process 0, then write it to a
|
|
d@0
|
72 file.
|
|
d@0
|
73 <a name="index-MPI_005fIO-382"></a><a name="index-MPI_005fANY_005fSOURCE-383"></a>
|
|
d@0
|
74 Second, in principle we may want to have separate wisdom for every
|
|
d@0
|
75 process, since in general the processes may run on different hardware
|
|
d@0
|
76 even for a single MPI program. However, in practice FFTW's MPI code
|
|
d@0
|
77 is designed for the case of homogeneous hardware (see <a href="Load-balancing.html#Load-balancing">Load balancing</a>), and in this case it is convenient to use the same wisdom
|
|
d@0
|
78 for every process. Thus, we need a mechanism to synchronize the wisdom.
|
|
d@0
|
79
|
|
d@0
|
80 <p>To address both of these problems, FFTW provides the following two
|
|
d@0
|
81 functions:
|
|
d@0
|
82
|
|
d@0
|
83 <pre class="example"> void fftw_mpi_broadcast_wisdom(MPI_Comm comm);
|
|
d@0
|
84 void fftw_mpi_gather_wisdom(MPI_Comm comm);
|
|
d@0
|
85 </pre>
|
|
d@0
|
86 <p><a name="index-fftw_005fmpi_005fgather_005fwisdom-384"></a><a name="index-fftw_005fmpi_005fbroadcast_005fwisdom-385"></a>
|
|
d@0
|
87 Given a communicator <code>comm</code>, <code>fftw_mpi_broadcast_wisdom</code>
|
|
d@0
|
88 will broadcast the wisdom from process 0 to all other processes.
|
|
d@0
|
89 Conversely, <code>fftw_mpi_gather_wisdom</code> will collect wisdom from all
|
|
d@0
|
90 processes onto process 0. (If the plans created for the same problem
|
|
d@0
|
91 by different processes are not the same, <code>fftw_mpi_gather_wisdom</code>
|
|
d@0
|
92 will arbitrarily choose one of the plans.) Both of these functions
|
|
d@0
|
93 may result in suboptimal plans for different processes if the
|
|
d@0
|
94 processes are running on non-identical hardware. Both of these
|
|
d@0
|
95 functions are <em>collective</em> calls, which means that they must be
|
|
d@0
|
96 executed by all processes in the communicator.
|
|
d@0
|
97 <a name="index-collective-function-386"></a>
|
|
d@0
|
98 So, for example, a typical code snippet to import wisdom from a file
|
|
d@0
|
99 and use it on all processes would be:
|
|
d@0
|
100
|
|
d@0
|
101 <pre class="example"> {
|
|
d@0
|
102 int rank;
|
|
d@0
|
103 FILE *f;
|
|
d@0
|
104
|
|
d@0
|
105 fftw_mpi_init();
|
|
d@0
|
106 MPI_Comm_rank(MPI_COMM_WORLD, &rank);
|
|
d@0
|
107 if (rank == 0 && (f = fopen("mywisdom", "r"))) {
|
|
d@0
|
108 fftw_import_wisdom_from_file(f);
|
|
d@0
|
109 fclose(f);
|
|
d@0
|
110 }
|
|
d@0
|
111 fftw_mpi_broadcast_wisdom(MPI_COMM_WORLD);
|
|
d@0
|
112 }
|
|
d@0
|
113 </pre>
|
|
d@0
|
114 <p>(Note that we must call <code>fftw_mpi_init</code> before importing any
|
|
d@0
|
115 wisdom that might contain MPI plans.) Similarly, a typical code
|
|
d@0
|
116 snippet to export wisdom from all processes to a file is:
|
|
d@0
|
117 <a name="index-fftw_005fmpi_005finit-387"></a>
|
|
d@0
|
118 <pre class="example"> {
|
|
d@0
|
119 int rank;
|
|
d@0
|
120 FILE *f;
|
|
d@0
|
121
|
|
d@0
|
122 fftw_mpi_gather_wisdom(MPI_COMM_WORLD);
|
|
d@0
|
123 MPI_Comm_rank(MPI_COMM_WORLD, &rank);
|
|
d@0
|
124 if (rank == 0 && (f = fopen("mywisdom", "w"))) {
|
|
d@0
|
125 fftw_export_wisdom_to_file(f);
|
|
d@0
|
126 fclose(f);
|
|
d@0
|
127 }
|
|
d@0
|
128 }
|
|
d@0
|
129 </pre>
|
|
d@0
|
130 <!-- -->
|
|
d@0
|
131 </body></html>
|
|
d@0
|
132
|
