cannam@127
|
1 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
|
cannam@127
|
2 <html>
|
cannam@127
|
3 <!-- This manual is for FFTW
|
cannam@127
|
4 (version 3.3.5, 30 July 2016).
|
cannam@127
|
5
|
cannam@127
|
6 Copyright (C) 2003 Matteo Frigo.
|
cannam@127
|
7
|
cannam@127
|
8 Copyright (C) 2003 Massachusetts Institute of Technology.
|
cannam@127
|
9
|
cannam@127
|
10 Permission is granted to make and distribute verbatim copies of this
|
cannam@127
|
11 manual provided the copyright notice and this permission notice are
|
cannam@127
|
12 preserved on all copies.
|
cannam@127
|
13
|
cannam@127
|
14 Permission is granted to copy and distribute modified versions of this
|
cannam@127
|
15 manual under the conditions for verbatim copying, provided that the
|
cannam@127
|
16 entire resulting derived work is distributed under the terms of a
|
cannam@127
|
17 permission notice identical to this one.
|
cannam@127
|
18
|
cannam@127
|
19 Permission is granted to copy and distribute translations of this manual
|
cannam@127
|
20 into another language, under the above conditions for modified versions,
|
cannam@127
|
21 except that this permission notice may be stated in a translation
|
cannam@127
|
22 approved by the Free Software Foundation. -->
|
cannam@127
|
23 <!-- Created by GNU Texinfo 5.2, http://www.gnu.org/software/texinfo/ -->
|
cannam@127
|
24 <head>
|
cannam@127
|
25 <title>FFTW 3.3.5: 2d MPI example</title>
|
cannam@127
|
26
|
cannam@127
|
27 <meta name="description" content="FFTW 3.3.5: 2d MPI example">
|
cannam@127
|
28 <meta name="keywords" content="FFTW 3.3.5: 2d MPI example">
|
cannam@127
|
29 <meta name="resource-type" content="document">
|
cannam@127
|
30 <meta name="distribution" content="global">
|
cannam@127
|
31 <meta name="Generator" content="makeinfo">
|
cannam@127
|
32 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
|
cannam@127
|
33 <link href="index.html#Top" rel="start" title="Top">
|
cannam@127
|
34 <link href="Concept-Index.html#Concept-Index" rel="index" title="Concept Index">
|
cannam@127
|
35 <link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
|
cannam@127
|
36 <link href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" rel="up" title="Distributed-memory FFTW with MPI">
|
cannam@127
|
37 <link href="MPI-Data-Distribution.html#MPI-Data-Distribution" rel="next" title="MPI Data Distribution">
|
cannam@127
|
38 <link href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW" rel="prev" title="Linking and Initializing MPI FFTW">
|
cannam@127
|
39 <style type="text/css">
|
cannam@127
|
40 <!--
|
cannam@127
|
41 a.summary-letter {text-decoration: none}
|
cannam@127
|
42 blockquote.smallquotation {font-size: smaller}
|
cannam@127
|
43 div.display {margin-left: 3.2em}
|
cannam@127
|
44 div.example {margin-left: 3.2em}
|
cannam@127
|
45 div.indentedblock {margin-left: 3.2em}
|
cannam@127
|
46 div.lisp {margin-left: 3.2em}
|
cannam@127
|
47 div.smalldisplay {margin-left: 3.2em}
|
cannam@127
|
48 div.smallexample {margin-left: 3.2em}
|
cannam@127
|
49 div.smallindentedblock {margin-left: 3.2em; font-size: smaller}
|
cannam@127
|
50 div.smalllisp {margin-left: 3.2em}
|
cannam@127
|
51 kbd {font-style:oblique}
|
cannam@127
|
52 pre.display {font-family: inherit}
|
cannam@127
|
53 pre.format {font-family: inherit}
|
cannam@127
|
54 pre.menu-comment {font-family: serif}
|
cannam@127
|
55 pre.menu-preformatted {font-family: serif}
|
cannam@127
|
56 pre.smalldisplay {font-family: inherit; font-size: smaller}
|
cannam@127
|
57 pre.smallexample {font-size: smaller}
|
cannam@127
|
58 pre.smallformat {font-family: inherit; font-size: smaller}
|
cannam@127
|
59 pre.smalllisp {font-size: smaller}
|
cannam@127
|
60 span.nocodebreak {white-space:nowrap}
|
cannam@127
|
61 span.nolinebreak {white-space:nowrap}
|
cannam@127
|
62 span.roman {font-family:serif; font-weight:normal}
|
cannam@127
|
63 span.sansserif {font-family:sans-serif; font-weight:normal}
|
cannam@127
|
64 ul.no-bullet {list-style: none}
|
cannam@127
|
65 -->
|
cannam@127
|
66 </style>
|
cannam@127
|
67
|
cannam@127
|
68
|
cannam@127
|
69 </head>
|
cannam@127
|
70
|
cannam@127
|
71 <body lang="en" bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#800080" alink="#FF0000">
|
cannam@127
|
72 <a name="g_t2d-MPI-example"></a>
|
cannam@127
|
73 <div class="header">
|
cannam@127
|
74 <p>
|
cannam@127
|
75 Next: <a href="MPI-Data-Distribution.html#MPI-Data-Distribution" accesskey="n" rel="next">MPI Data Distribution</a>, Previous: <a href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW" accesskey="p" rel="prev">Linking and Initializing MPI FFTW</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
|
cannam@127
|
76 </div>
|
cannam@127
|
77 <hr>
|
cannam@127
|
78 <a name="g_t2d-MPI-example-1"></a>
|
cannam@127
|
79 <h3 class="section">6.3 2d MPI example</h3>
|
cannam@127
|
80
|
cannam@127
|
81 <p>Before we document the FFTW MPI interface in detail, we begin with a
|
cannam@127
|
82 simple example outlining how one would perform a two-dimensional
|
cannam@127
|
83 <code>N0</code> by <code>N1</code> complex DFT.
|
cannam@127
|
84 </p>
|
cannam@127
|
85 <div class="example">
|
cannam@127
|
86 <pre class="example">#include <fftw3-mpi.h>
|
cannam@127
|
87
|
cannam@127
|
88 int main(int argc, char **argv)
|
cannam@127
|
89 {
|
cannam@127
|
90 const ptrdiff_t N0 = ..., N1 = ...;
|
cannam@127
|
91 fftw_plan plan;
|
cannam@127
|
92 fftw_complex *data;
|
cannam@127
|
93 ptrdiff_t alloc_local, local_n0, local_0_start, i, j;
|
cannam@127
|
94
|
cannam@127
|
95 MPI_Init(&argc, &argv);
|
cannam@127
|
96 fftw_mpi_init();
|
cannam@127
|
97
|
cannam@127
|
98 /* <span class="roman">get local data size and allocate</span> */
|
cannam@127
|
99 alloc_local = fftw_mpi_local_size_2d(N0, N1, MPI_COMM_WORLD,
|
cannam@127
|
100 &local_n0, &local_0_start);
|
cannam@127
|
101 data = fftw_alloc_complex(alloc_local);
|
cannam@127
|
102
|
cannam@127
|
103 /* <span class="roman">create plan for in-place forward DFT</span> */
|
cannam@127
|
104 plan = fftw_mpi_plan_dft_2d(N0, N1, data, data, MPI_COMM_WORLD,
|
cannam@127
|
105 FFTW_FORWARD, FFTW_ESTIMATE);
|
cannam@127
|
106
|
cannam@127
|
107 /* <span class="roman">initialize data to some function</span> my_function(x,y) */
|
cannam@127
|
108 for (i = 0; i < local_n0; ++i) for (j = 0; j < N1; ++j)
|
cannam@127
|
109 data[i*N1 + j] = my_function(local_0_start + i, j);
|
cannam@127
|
110
|
cannam@127
|
111 /* <span class="roman">compute transforms, in-place, as many times as desired</span> */
|
cannam@127
|
112 fftw_execute(plan);
|
cannam@127
|
113
|
cannam@127
|
114 fftw_destroy_plan(plan);
|
cannam@127
|
115
|
cannam@127
|
116 MPI_Finalize();
|
cannam@127
|
117 }
|
cannam@127
|
118 </pre></div>
|
cannam@127
|
119
|
cannam@127
|
120 <p>As can be seen above, the MPI interface follows the same basic style
|
cannam@127
|
121 of allocate/plan/execute/destroy as the serial FFTW routines. All of
|
cannam@127
|
122 the MPI-specific routines are prefixed with ‘<samp>fftw_mpi_</samp>’ instead
|
cannam@127
|
123 of ‘<samp>fftw_</samp>’. There are a few important differences, however:
|
cannam@127
|
124 </p>
|
cannam@127
|
125 <p>First, we must call <code>fftw_mpi_init()</code> after calling
|
cannam@127
|
126 <code>MPI_Init</code> (required in all MPI programs) and before calling any
|
cannam@127
|
127 other ‘<samp>fftw_mpi_</samp>’ routine.
|
cannam@127
|
128 <a name="index-MPI_005fInit"></a>
|
cannam@127
|
129 <a name="index-fftw_005fmpi_005finit-1"></a>
|
cannam@127
|
130 </p>
|
cannam@127
|
131
|
cannam@127
|
132 <p>Second, when we create the plan with <code>fftw_mpi_plan_dft_2d</code>,
|
cannam@127
|
133 analogous to <code>fftw_plan_dft_2d</code>, we pass an additional argument:
|
cannam@127
|
134 the communicator, indicating which processes will participate in the
|
cannam@127
|
135 transform (here <code>MPI_COMM_WORLD</code>, indicating all processes).
|
cannam@127
|
136 Whenever you create, execute, or destroy a plan for an MPI transform,
|
cannam@127
|
137 you must call the corresponding FFTW routine on <em>all</em> processes
|
cannam@127
|
138 in the communicator for that transform. (That is, these are
|
cannam@127
|
139 <em>collective</em> calls.) Note that the plan for the MPI transform
|
cannam@127
|
140 uses the standard <code>fftw_execute</code> and <code>fftw_destroy</code> routines
|
cannam@127
|
141 (on the other hand, there are MPI-specific new-array execute functions
|
cannam@127
|
142 documented below).
|
cannam@127
|
143 <a name="index-collective-function"></a>
|
cannam@127
|
144 <a name="index-fftw_005fmpi_005fplan_005fdft_005f2d"></a>
|
cannam@127
|
145 <a name="index-MPI_005fCOMM_005fWORLD-1"></a>
|
cannam@127
|
146 </p>
|
cannam@127
|
147
|
cannam@127
|
148 <p>Third, all of the FFTW MPI routines take <code>ptrdiff_t</code> arguments
|
cannam@127
|
149 instead of <code>int</code> as for the serial FFTW. <code>ptrdiff_t</code> is a
|
cannam@127
|
150 standard C integer type which is (at least) 32 bits wide on a 32-bit
|
cannam@127
|
151 machine and 64 bits wide on a 64-bit machine. This is to make it easy
|
cannam@127
|
152 to specify very large parallel transforms on a 64-bit machine. (You
|
cannam@127
|
153 can specify 64-bit transform sizes in the serial FFTW, too, but only
|
cannam@127
|
154 by using the ‘<samp>guru64</samp>’ planner interface. See <a href="64_002dbit-Guru-Interface.html#g_t64_002dbit-Guru-Interface">64-bit Guru Interface</a>.)
|
cannam@127
|
155 <a name="index-ptrdiff_005ft-1"></a>
|
cannam@127
|
156 <a name="index-64_002dbit-architecture-1"></a>
|
cannam@127
|
157 </p>
|
cannam@127
|
158
|
cannam@127
|
159 <p>Fourth, and most importantly, you don’t allocate the entire
|
cannam@127
|
160 two-dimensional array on each process. Instead, you call
|
cannam@127
|
161 <code>fftw_mpi_local_size_2d</code> to find out what <em>portion</em> of the
|
cannam@127
|
162 array resides on each processor, and how much space to allocate.
|
cannam@127
|
163 Here, the portion of the array on each process is a <code>local_n0</code> by
|
cannam@127
|
164 <code>N1</code> slice of the total array, starting at index
|
cannam@127
|
165 <code>local_0_start</code>. The total number of <code>fftw_complex</code> numbers
|
cannam@127
|
166 to allocate is given by the <code>alloc_local</code> return value, which
|
cannam@127
|
167 <em>may</em> be greater than <code>local_n0 * N1</code> (in case some
|
cannam@127
|
168 intermediate calculations require additional storage). The data
|
cannam@127
|
169 distribution in FFTW’s MPI interface is described in more detail by
|
cannam@127
|
170 the next section.
|
cannam@127
|
171 <a name="index-fftw_005fmpi_005flocal_005fsize_005f2d"></a>
|
cannam@127
|
172 <a name="index-data-distribution-1"></a>
|
cannam@127
|
173 </p>
|
cannam@127
|
174
|
cannam@127
|
175 <p>Given the portion of the array that resides on the local process, it
|
cannam@127
|
176 is straightforward to initialize the data (here to a function
|
cannam@127
|
177 <code>myfunction</code>) and otherwise manipulate it. Of course, at the end
|
cannam@127
|
178 of the program you may want to output the data somehow, but
|
cannam@127
|
179 synchronizing this output is up to you and is beyond the scope of this
|
cannam@127
|
180 manual. (One good way to output a large multi-dimensional distributed
|
cannam@127
|
181 array in MPI to a portable binary file is to use the free HDF5
|
cannam@127
|
182 library; see the <a href="http://www.hdfgroup.org/">HDF home page</a>.)
|
cannam@127
|
183 <a name="index-HDF5"></a>
|
cannam@127
|
184 <a name="index-MPI-I_002fO"></a>
|
cannam@127
|
185 </p>
|
cannam@127
|
186 <hr>
|
cannam@127
|
187 <div class="header">
|
cannam@127
|
188 <p>
|
cannam@127
|
189 Next: <a href="MPI-Data-Distribution.html#MPI-Data-Distribution" accesskey="n" rel="next">MPI Data Distribution</a>, Previous: <a href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW" accesskey="p" rel="prev">Linking and Initializing MPI FFTW</a>, Up: <a href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" accesskey="u" rel="up">Distributed-memory FFTW with MPI</a> [<a href="index.html#SEC_Contents" title="Table of contents" rel="contents">Contents</a>][<a href="Concept-Index.html#Concept-Index" title="Index" rel="index">Index</a>]</p>
|
cannam@127
|
190 </div>
|
cannam@127
|
191
|
cannam@127
|
192
|
cannam@127
|
193
|
cannam@127
|
194 </body>
|
cannam@127
|
195 </html>
|