Mercurial > hg > sv-dependency-builds

annotate src/fftw-3.3.3/doc/html/2d-MPI-example.html @ 10:37bf6b4a2645

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Add FFTW3
author Chris Cannam
date Wed, 20 Mar 2013 15:35:50 +0000
parents
children
rev   line source
Chris@10 1 <html lang="en">
Chris@10 2 <head>
Chris@10 3 <title>2d MPI example - FFTW 3.3.3</title>
Chris@10 4 <meta http-equiv="Content-Type" content="text/html">
Chris@10 5 <meta name="description" content="FFTW 3.3.3">
Chris@10 6 <meta name="generator" content="makeinfo 4.13">
Chris@10 7 <link title="Top" rel="start" href="index.html#Top">
Chris@10 8 <link rel="up" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI" title="Distributed-memory FFTW with MPI">
Chris@10 9 <link rel="prev" href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW" title="Linking and Initializing MPI FFTW">
Chris@10 10 <link rel="next" href="MPI-Data-Distribution.html#MPI-Data-Distribution" title="MPI Data Distribution">
Chris@10 11 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
Chris@10 12 <!--
Chris@10 13 This manual is for FFTW
Chris@10 14 (version 3.3.3, 25 November 2012).
Chris@10 15
Chris@10 16 Copyright (C) 2003 Matteo Frigo.
Chris@10 17
Chris@10 18 Copyright (C) 2003 Massachusetts Institute of Technology.
Chris@10 19
Chris@10 20 Permission is granted to make and distribute verbatim copies of
Chris@10 21 this manual provided the copyright notice and this permission
Chris@10 22 notice are preserved on all copies.
Chris@10 23
Chris@10 24 Permission is granted to copy and distribute modified versions of
Chris@10 25 this manual under the conditions for verbatim copying, provided
Chris@10 26 that the entire resulting derived work is distributed under the
Chris@10 27 terms of a permission notice identical to this one.
Chris@10 28
Chris@10 29 Permission is granted to copy and distribute translations of this
Chris@10 30 manual into another language, under the above conditions for
Chris@10 31 modified versions, except that this permission notice may be
Chris@10 32 stated in a translation approved by the Free Software Foundation.
Chris@10 33 -->
Chris@10 34 <meta http-equiv="Content-Style-Type" content="text/css">
Chris@10 35 <style type="text/css"><!--
Chris@10 36 pre.display { font-family:inherit }
Chris@10 37 pre.format { font-family:inherit }
Chris@10 38 pre.smalldisplay { font-family:inherit; font-size:smaller }
Chris@10 39 pre.smallformat { font-family:inherit; font-size:smaller }
Chris@10 40 pre.smallexample { font-size:smaller }
Chris@10 41 pre.smalllisp { font-size:smaller }
Chris@10 42 span.sc { font-variant:small-caps }
Chris@10 43 span.roman { font-family:serif; font-weight:normal; }
Chris@10 44 span.sansserif { font-family:sans-serif; font-weight:normal; }
Chris@10 45 --></style>
Chris@10 46 </head>
Chris@10 47 <body>
Chris@10 48 <div class="node">
Chris@10 49 <a name="g_t2d-MPI-example"></a>
Chris@10 50 <p>
Chris@10 51 Next:&nbsp;<a rel="next" accesskey="n" href="MPI-Data-Distribution.html#MPI-Data-Distribution">MPI Data Distribution</a>,
Chris@10 52 Previous:&nbsp;<a rel="previous" accesskey="p" href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW">Linking and Initializing MPI FFTW</a>,
Chris@10 53 Up:&nbsp;<a rel="up" accesskey="u" href="Distributed_002dmemory-FFTW-with-MPI.html#Distributed_002dmemory-FFTW-with-MPI">Distributed-memory FFTW with MPI</a>
Chris@10 54 <hr>
Chris@10 55 </div>
Chris@10 56
Chris@10 57 <h3 class="section">6.3 2d MPI example</h3>
Chris@10 58
Chris@10 59 <p>Before we document the FFTW MPI interface in detail, we begin with a
Chris@10 60 simple example outlining how one would perform a two-dimensional
Chris@10 61 <code>N0</code> by <code>N1</code> complex DFT.
Chris@10 62
Chris@10 63 <pre class="example"> #include &lt;fftw3-mpi.h&gt;
Chris@10 64
Chris@10 65 int main(int argc, char **argv)
Chris@10 66 {
Chris@10 67 const ptrdiff_t N0 = ..., N1 = ...;
Chris@10 68 fftw_plan plan;
Chris@10 69 fftw_complex *data;
Chris@10 70 ptrdiff_t alloc_local, local_n0, local_0_start, i, j;
Chris@10 71
Chris@10 72 MPI_Init(&amp;argc, &amp;argv);
Chris@10 73 fftw_mpi_init();
Chris@10 74
Chris@10 75 /* <span class="roman">get local data size and allocate</span> */
Chris@10 76 alloc_local = fftw_mpi_local_size_2d(N0, N1, MPI_COMM_WORLD,
Chris@10 77 &amp;local_n0, &amp;local_0_start);
Chris@10 78 data = fftw_alloc_complex(alloc_local);
Chris@10 79
Chris@10 80 /* <span class="roman">create plan for in-place forward DFT</span> */
Chris@10 81 plan = fftw_mpi_plan_dft_2d(N0, N1, data, data, MPI_COMM_WORLD,
Chris@10 82 FFTW_FORWARD, FFTW_ESTIMATE);
Chris@10 83
Chris@10 84 /* <span class="roman">initialize data to some function</span> my_function(x,y) */
Chris@10 85 for (i = 0; i &lt; local_n0; ++i) for (j = 0; j &lt; N1; ++j)
Chris@10 86 data[i*N1 + j] = my_function(local_0_start + i, j);
Chris@10 87
Chris@10 88 /* <span class="roman">compute transforms, in-place, as many times as desired</span> */
Chris@10 89 fftw_execute(plan);
Chris@10 90
Chris@10 91 fftw_destroy_plan(plan);
Chris@10 92
Chris@10 93 MPI_Finalize();
Chris@10 94 }
Chris@10 95 </pre>
Chris@10 96 <p>As can be seen above, the MPI interface follows the same basic style
Chris@10 97 of allocate/plan/execute/destroy as the serial FFTW routines. All of
Chris@10 98 the MPI-specific routines are prefixed with &lsquo;<samp><span class="samp">fftw_mpi_</span></samp>&rsquo; instead
Chris@10 99 of &lsquo;<samp><span class="samp">fftw_</span></samp>&rsquo;. There are a few important differences, however:
Chris@10 100
Chris@10 101 <p>First, we must call <code>fftw_mpi_init()</code> after calling
Chris@10 102 <code>MPI_Init</code> (required in all MPI programs) and before calling any
Chris@10 103 other &lsquo;<samp><span class="samp">fftw_mpi_</span></samp>&rsquo; routine.
Chris@10 104 <a name="index-MPI_005fInit-357"></a><a name="index-fftw_005fmpi_005finit-358"></a>
Chris@10 105
Chris@10 106 <p>Second, when we create the plan with <code>fftw_mpi_plan_dft_2d</code>,
Chris@10 107 analogous to <code>fftw_plan_dft_2d</code>, we pass an additional argument:
Chris@10 108 the communicator, indicating which processes will participate in the
Chris@10 109 transform (here <code>MPI_COMM_WORLD</code>, indicating all processes).
Chris@10 110 Whenever you create, execute, or destroy a plan for an MPI transform,
Chris@10 111 you must call the corresponding FFTW routine on <em>all</em> processes
Chris@10 112 in the communicator for that transform. (That is, these are
Chris@10 113 <em>collective</em> calls.) Note that the plan for the MPI transform
Chris@10 114 uses the standard <code>fftw_execute</code> and <code>fftw_destroy</code> routines
Chris@10 115 (on the other hand, there are MPI-specific new-array execute functions
Chris@10 116 documented below).
Chris@10 117 <a name="index-collective-function-359"></a><a name="index-fftw_005fmpi_005fplan_005fdft_005f2d-360"></a><a name="index-MPI_005fCOMM_005fWORLD-361"></a>
Chris@10 118
Chris@10 119 <p>Third, all of the FFTW MPI routines take <code>ptrdiff_t</code> arguments
Chris@10 120 instead of <code>int</code> as for the serial FFTW. <code>ptrdiff_t</code> is a
Chris@10 121 standard C integer type which is (at least) 32 bits wide on a 32-bit
Chris@10 122 machine and 64 bits wide on a 64-bit machine. This is to make it easy
Chris@10 123 to specify very large parallel transforms on a 64-bit machine. (You
Chris@10 124 can specify 64-bit transform sizes in the serial FFTW, too, but only
Chris@10 125 by using the &lsquo;<samp><span class="samp">guru64</span></samp>&rsquo; planner interface. See <a href="64_002dbit-Guru-Interface.html#g_t64_002dbit-Guru-Interface">64-bit Guru Interface</a>.)
Chris@10 126 <a name="index-ptrdiff_005ft-362"></a><a name="index-g_t64_002dbit-architecture-363"></a>
Chris@10 127
Chris@10 128 <p>Fourth, and most importantly, you don't allocate the entire
Chris@10 129 two-dimensional array on each process. Instead, you call
Chris@10 130 <code>fftw_mpi_local_size_2d</code> to find out what <em>portion</em> of the
Chris@10 131 array resides on each processor, and how much space to allocate.
Chris@10 132 Here, the portion of the array on each process is a <code>local_n0</code> by
Chris@10 133 <code>N1</code> slice of the total array, starting at index
Chris@10 134 <code>local_0_start</code>. The total number of <code>fftw_complex</code> numbers
Chris@10 135 to allocate is given by the <code>alloc_local</code> return value, which
Chris@10 136 <em>may</em> be greater than <code>local_n0 * N1</code> (in case some
Chris@10 137 intermediate calculations require additional storage). The data
Chris@10 138 distribution in FFTW's MPI interface is described in more detail by
Chris@10 139 the next section.
Chris@10 140 <a name="index-fftw_005fmpi_005flocal_005fsize_005f2d-364"></a><a name="index-data-distribution-365"></a>
Chris@10 141
Chris@10 142 <p>Given the portion of the array that resides on the local process, it
Chris@10 143 is straightforward to initialize the data (here to a function
Chris@10 144 <code>myfunction</code>) and otherwise manipulate it. Of course, at the end
Chris@10 145 of the program you may want to output the data somehow, but
Chris@10 146 synchronizing this output is up to you and is beyond the scope of this
Chris@10 147 manual. (One good way to output a large multi-dimensional distributed
Chris@10 148 array in MPI to a portable binary file is to use the free HDF5
Chris@10 149 library; see the <a href="http://www.hdfgroup.org/">HDF home page</a>.)
Chris@10 150 <a name="index-HDF5-366"></a><a name="index-MPI-I_002fO-367"></a>
Chris@10 151 <!-- -->
Chris@10 152
Chris@10 153 </body></html>
Chris@10 154