annotate src/fftw-3.3.8/doc/html/2d-MPI-example.html @ 167:bd3cc4d1df30

Add FFTW 3.3.8 source, and a Linux build
author Chris Cannam <cannam@all-day-breakfast.com>
date Tue, 19 Nov 2019 14:52:55 +0000
parents
children
rev   line source
cannam@167 1 <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
cannam@167 2 <html>
cannam@167 3 <!-- This manual is for FFTW
cannam@167 4 (version 3.3.8, 24 May 2018).
cannam@167 5
cannam@167 6 Copyright (C) 2003 Matteo Frigo.
cannam@167 7
cannam@167 8 Copyright (C) 2003 Massachusetts Institute of Technology.
cannam@167 9
cannam@167 10 Permission is granted to make and distribute verbatim copies of this
cannam@167 11 manual provided the copyright notice and this permission notice are
cannam@167 12 preserved on all copies.
cannam@167 13
cannam@167 14 Permission is granted to copy and distribute modified versions of this
cannam@167 15 manual under the conditions for verbatim copying, provided that the
cannam@167 16 entire resulting derived work is distributed under the terms of a
cannam@167 17 permission notice identical to this one.
cannam@167 18
cannam@167 19 Permission is granted to copy and distribute translations of this manual
cannam@167 20 into another language, under the above conditions for modified versions,
cannam@167 21 except that this permission notice may be stated in a translation
cannam@167 22 approved by the Free Software Foundation. -->
cannam@167 23 <!-- Created by GNU Texinfo 6.3, http://www.gnu.org/software/texinfo/ -->
cannam@167 24 <head>
cannam@167 25 <title>FFTW 3.3.8: 2d MPI example</title>
cannam@167 26
cannam@167 27 <meta name="description" content="FFTW 3.3.8: 2d MPI example">
cannam@167 28 <meta name="keywords" content="FFTW 3.3.8: 2d MPI example">
cannam@167 29 <meta name="resource-type" content="document">
cannam@167 30 <meta name="distribution" content="global">
cannam@167 31 <meta name="Generator" content="makeinfo">
cannam@167 32 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
cannam@167 33 <link href="index.html#Top" rel="start" title="Top">
cannam@167 34 <link href="Concept-Index.html#Concept-Index" rel="index" title="Concept Index">
cannam@167 35 <link href="index.html#SEC_Contents" rel="contents" title="Table of Contents">
cannam@167 36 <link href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" rel="up" title="Distributed-memory FFTW with MPI">
cannam@167 37 <link href="MPI-Data-Distribution.html#MPI-Data-Distribution" rel="next" title="MPI Data Distribution">
cannam@167 38 <link href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW" rel="prev" title="Linking and Initializing MPI FFTW">
cannam@167 39 <style type="text/css">
cannam@167 40 <!--
cannam@167 41 a.summary-letter {text-decoration: none}
cannam@167 42 blockquote.indentedblock {margin-right: 0em}
cannam@167 43 blockquote.smallindentedblock {margin-right: 0em; font-size: smaller}
cannam@167 44 blockquote.smallquotation {font-size: smaller}
cannam@167 45 div.display {margin-left: 3.2em}
cannam@167 46 div.example {margin-left: 3.2em}
cannam@167 47 div.lisp {margin-left: 3.2em}
cannam@167 48 div.smalldisplay {margin-left: 3.2em}
cannam@167 49 div.smallexample {margin-left: 3.2em}
cannam@167 50 div.smalllisp {margin-left: 3.2em}
cannam@167 51 kbd {font-style: oblique}
cannam@167 52 pre.display {font-family: inherit}
cannam@167 53 pre.format {font-family: inherit}
cannam@167 54 pre.menu-comment {font-family: serif}
cannam@167 55 pre.menu-preformatted {font-family: serif}
cannam@167 56 pre.smalldisplay {font-family: inherit; font-size: smaller}
cannam@167 57 pre.smallexample {font-size: smaller}
cannam@167 58 pre.smallformat {font-family: inherit; font-size: smaller}
cannam@167 59 pre.smalllisp {font-size: smaller}
cannam@167 60 span.nolinebreak {white-space: nowrap}
cannam@167 61 span.roman {font-family: initial; font-weight: normal}
cannam@167 62 span.sansserif {font-family: sans-serif; font-weight: normal}
cannam@167 63 ul.no-bullet {list-style: none}
cannam@167 64 -->
cannam@167 65 </style>
cannam@167 66
cannam@167 67
cannam@167 68 </head>
cannam@167 69
cannam@167 70 <body lang="en">
cannam@167 71 <a name="g_t2d-MPI-example"></a>
cannam@167 72 <div class="header">
cannam@167 73 <p>
cannam@167 74 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> &nbsp; [<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@167 75 </div>
cannam@167 76 <hr>
cannam@167 77 <a name="g_t2d-MPI-example-1"></a>
cannam@167 78 <h3 class="section">6.3 2d MPI example</h3>
cannam@167 79
cannam@167 80 <p>Before we document the FFTW MPI interface in detail, we begin with a
cannam@167 81 simple example outlining how one would perform a two-dimensional
cannam@167 82 <code>N0</code> by <code>N1</code> complex DFT.
cannam@167 83 </p>
cannam@167 84 <div class="example">
cannam@167 85 <pre class="example">#include &lt;fftw3-mpi.h&gt;
cannam@167 86
cannam@167 87 int main(int argc, char **argv)
cannam@167 88 {
cannam@167 89 const ptrdiff_t N0 = ..., N1 = ...;
cannam@167 90 fftw_plan plan;
cannam@167 91 fftw_complex *data;
cannam@167 92 ptrdiff_t alloc_local, local_n0, local_0_start, i, j;
cannam@167 93
cannam@167 94 MPI_Init(&amp;argc, &amp;argv);
cannam@167 95 fftw_mpi_init();
cannam@167 96
cannam@167 97 /* <span class="roman">get local data size and allocate</span> */
cannam@167 98 alloc_local = fftw_mpi_local_size_2d(N0, N1, MPI_COMM_WORLD,
cannam@167 99 &amp;local_n0, &amp;local_0_start);
cannam@167 100 data = fftw_alloc_complex(alloc_local);
cannam@167 101
cannam@167 102 /* <span class="roman">create plan for in-place forward DFT</span> */
cannam@167 103 plan = fftw_mpi_plan_dft_2d(N0, N1, data, data, MPI_COMM_WORLD,
cannam@167 104 FFTW_FORWARD, FFTW_ESTIMATE);
cannam@167 105
cannam@167 106 /* <span class="roman">initialize data to some function</span> my_function(x,y) */
cannam@167 107 for (i = 0; i &lt; local_n0; ++i) for (j = 0; j &lt; N1; ++j)
cannam@167 108 data[i*N1 + j] = my_function(local_0_start + i, j);
cannam@167 109
cannam@167 110 /* <span class="roman">compute transforms, in-place, as many times as desired</span> */
cannam@167 111 fftw_execute(plan);
cannam@167 112
cannam@167 113 fftw_destroy_plan(plan);
cannam@167 114
cannam@167 115 MPI_Finalize();
cannam@167 116 }
cannam@167 117 </pre></div>
cannam@167 118
cannam@167 119 <p>As can be seen above, the MPI interface follows the same basic style
cannam@167 120 of allocate/plan/execute/destroy as the serial FFTW routines. All of
cannam@167 121 the MPI-specific routines are prefixed with &lsquo;<samp>fftw_mpi_</samp>&rsquo; instead
cannam@167 122 of &lsquo;<samp>fftw_</samp>&rsquo;. There are a few important differences, however:
cannam@167 123 </p>
cannam@167 124 <p>First, we must call <code>fftw_mpi_init()</code> after calling
cannam@167 125 <code>MPI_Init</code> (required in all MPI programs) and before calling any
cannam@167 126 other &lsquo;<samp>fftw_mpi_</samp>&rsquo; routine.
cannam@167 127 <a name="index-MPI_005fInit"></a>
cannam@167 128 <a name="index-fftw_005fmpi_005finit-1"></a>
cannam@167 129 </p>
cannam@167 130
cannam@167 131 <p>Second, when we create the plan with <code>fftw_mpi_plan_dft_2d</code>,
cannam@167 132 analogous to <code>fftw_plan_dft_2d</code>, we pass an additional argument:
cannam@167 133 the communicator, indicating which processes will participate in the
cannam@167 134 transform (here <code>MPI_COMM_WORLD</code>, indicating all processes).
cannam@167 135 Whenever you create, execute, or destroy a plan for an MPI transform,
cannam@167 136 you must call the corresponding FFTW routine on <em>all</em> processes
cannam@167 137 in the communicator for that transform. (That is, these are
cannam@167 138 <em>collective</em> calls.) Note that the plan for the MPI transform
cannam@167 139 uses the standard <code>fftw_execute</code> and <code>fftw_destroy</code> routines
cannam@167 140 (on the other hand, there are MPI-specific new-array execute functions
cannam@167 141 documented below).
cannam@167 142 <a name="index-collective-function"></a>
cannam@167 143 <a name="index-fftw_005fmpi_005fplan_005fdft_005f2d"></a>
cannam@167 144 <a name="index-MPI_005fCOMM_005fWORLD-1"></a>
cannam@167 145 </p>
cannam@167 146
cannam@167 147 <p>Third, all of the FFTW MPI routines take <code>ptrdiff_t</code> arguments
cannam@167 148 instead of <code>int</code> as for the serial FFTW. <code>ptrdiff_t</code> is a
cannam@167 149 standard C integer type which is (at least) 32 bits wide on a 32-bit
cannam@167 150 machine and 64 bits wide on a 64-bit machine. This is to make it easy
cannam@167 151 to specify very large parallel transforms on a 64-bit machine. (You
cannam@167 152 can specify 64-bit transform sizes in the serial FFTW, too, but only
cannam@167 153 by using the &lsquo;<samp>guru64</samp>&rsquo; planner interface. See <a href="64_002dbit-Guru-Interface.html#g_t64_002dbit-Guru-Interface">64-bit Guru Interface</a>.)
cannam@167 154 <a name="index-ptrdiff_005ft-1"></a>
cannam@167 155 <a name="index-64_002dbit-architecture-1"></a>
cannam@167 156 </p>
cannam@167 157
cannam@167 158 <p>Fourth, and most importantly, you don&rsquo;t allocate the entire
cannam@167 159 two-dimensional array on each process. Instead, you call
cannam@167 160 <code>fftw_mpi_local_size_2d</code> to find out what <em>portion</em> of the
cannam@167 161 array resides on each processor, and how much space to allocate.
cannam@167 162 Here, the portion of the array on each process is a <code>local_n0</code> by
cannam@167 163 <code>N1</code> slice of the total array, starting at index
cannam@167 164 <code>local_0_start</code>. The total number of <code>fftw_complex</code> numbers
cannam@167 165 to allocate is given by the <code>alloc_local</code> return value, which
cannam@167 166 <em>may</em> be greater than <code>local_n0 * N1</code> (in case some
cannam@167 167 intermediate calculations require additional storage). The data
cannam@167 168 distribution in FFTW&rsquo;s MPI interface is described in more detail by
cannam@167 169 the next section.
cannam@167 170 <a name="index-fftw_005fmpi_005flocal_005fsize_005f2d"></a>
cannam@167 171 <a name="index-data-distribution-1"></a>
cannam@167 172 </p>
cannam@167 173
cannam@167 174 <p>Given the portion of the array that resides on the local process, it
cannam@167 175 is straightforward to initialize the data (here to a function
cannam@167 176 <code>myfunction</code>) and otherwise manipulate it. Of course, at the end
cannam@167 177 of the program you may want to output the data somehow, but
cannam@167 178 synchronizing this output is up to you and is beyond the scope of this
cannam@167 179 manual. (One good way to output a large multi-dimensional distributed
cannam@167 180 array in MPI to a portable binary file is to use the free HDF5
cannam@167 181 library; see the <a href="http://www.hdfgroup.org/">HDF home page</a>.)
cannam@167 182 <a name="index-HDF5"></a>
cannam@167 183 <a name="index-MPI-I_002fO"></a>
cannam@167 184 </p>
cannam@167 185 <hr>
cannam@167 186 <div class="header">
cannam@167 187 <p>
cannam@167 188 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> &nbsp; [<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@167 189 </div>
cannam@167 190
cannam@167 191
cannam@167 192
cannam@167 193 </body>
cannam@167 194 </html>