|
cannam@95
|
1 <html lang="en">
|
|
cannam@95
|
2 <head>
|
|
cannam@95
|
3 <title>MPI Data Distribution Functions - FFTW 3.3.3</title>
|
|
cannam@95
|
4 <meta http-equiv="Content-Type" content="text/html">
|
|
cannam@95
|
5 <meta name="description" content="FFTW 3.3.3">
|
|
cannam@95
|
6 <meta name="generator" content="makeinfo 4.13">
|
|
cannam@95
|
7 <link title="Top" rel="start" href="index.html#Top">
|
|
cannam@95
|
8 <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference">
|
|
cannam@95
|
9 <link rel="prev" href="Using-MPI-Plans.html#Using-MPI-Plans" title="Using MPI Plans">
|
|
cannam@95
|
10 <link rel="next" href="MPI-Plan-Creation.html#MPI-Plan-Creation" title="MPI Plan Creation">
|
|
cannam@95
|
11 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
|
|
cannam@95
|
12 <!--
|
|
cannam@95
|
13 This manual is for FFTW
|
|
cannam@95
|
14 (version 3.3.3, 25 November 2012).
|
|
cannam@95
|
15
|
|
cannam@95
|
16 Copyright (C) 2003 Matteo Frigo.
|
|
cannam@95
|
17
|
|
cannam@95
|
18 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
cannam@95
|
19
|
|
cannam@95
|
20 Permission is granted to make and distribute verbatim copies of
|
|
cannam@95
|
21 this manual provided the copyright notice and this permission
|
|
cannam@95
|
22 notice are preserved on all copies.
|
|
cannam@95
|
23
|
|
cannam@95
|
24 Permission is granted to copy and distribute modified versions of
|
|
cannam@95
|
25 this manual under the conditions for verbatim copying, provided
|
|
cannam@95
|
26 that the entire resulting derived work is distributed under the
|
|
cannam@95
|
27 terms of a permission notice identical to this one.
|
|
cannam@95
|
28
|
|
cannam@95
|
29 Permission is granted to copy and distribute translations of this
|
|
cannam@95
|
30 manual into another language, under the above conditions for
|
|
cannam@95
|
31 modified versions, except that this permission notice may be
|
|
cannam@95
|
32 stated in a translation approved by the Free Software Foundation.
|
|
cannam@95
|
33 -->
|
|
cannam@95
|
34 <meta http-equiv="Content-Style-Type" content="text/css">
|
|
cannam@95
|
35 <style type="text/css"><!--
|
|
cannam@95
|
36 pre.display { font-family:inherit }
|
|
cannam@95
|
37 pre.format { font-family:inherit }
|
|
cannam@95
|
38 pre.smalldisplay { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
39 pre.smallformat { font-family:inherit; font-size:smaller }
|
|
cannam@95
|
40 pre.smallexample { font-size:smaller }
|
|
cannam@95
|
41 pre.smalllisp { font-size:smaller }
|
|
cannam@95
|
42 span.sc { font-variant:small-caps }
|
|
cannam@95
|
43 span.roman { font-family:serif; font-weight:normal; }
|
|
cannam@95
|
44 span.sansserif { font-family:sans-serif; font-weight:normal; }
|
|
cannam@95
|
45 --></style>
|
|
cannam@95
|
46 </head>
|
|
cannam@95
|
47 <body>
|
|
cannam@95
|
48 <div class="node">
|
|
cannam@95
|
49 <a name="MPI-Data-Distribution-Functions"></a>
|
|
cannam@95
|
50 <p>
|
|
cannam@95
|
51 Next: <a rel="next" accesskey="n" href="MPI-Plan-Creation.html#MPI-Plan-Creation">MPI Plan Creation</a>,
|
|
cannam@95
|
52 Previous: <a rel="previous" accesskey="p" href="Using-MPI-Plans.html#Using-MPI-Plans">Using MPI Plans</a>,
|
|
cannam@95
|
53 Up: <a rel="up" accesskey="u" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference">FFTW MPI Reference</a>
|
|
cannam@95
|
54 <hr>
|
|
cannam@95
|
55 </div>
|
|
cannam@95
|
56
|
|
cannam@95
|
57 <h4 class="subsection">6.12.4 MPI Data Distribution Functions</h4>
|
|
cannam@95
|
58
|
|
cannam@95
|
59 <p><a name="index-data-distribution-447"></a>As described above (see <a href="MPI-Data-Distribution.html#MPI-Data-Distribution">MPI Data Distribution</a>), in order to
|
|
cannam@95
|
60 allocate your arrays, <em>before</em> creating a plan, you must first
|
|
cannam@95
|
61 call one of the following routines to determine the required
|
|
cannam@95
|
62 allocation size and the portion of the array locally stored on a given
|
|
cannam@95
|
63 process. The <code>MPI_Comm</code> communicator passed here must be
|
|
cannam@95
|
64 equivalent to the communicator used below for plan creation.
|
|
cannam@95
|
65
|
|
cannam@95
|
66 <p>The basic interface for multidimensional transforms consists of the
|
|
cannam@95
|
67 functions:
|
|
cannam@95
|
68
|
|
cannam@95
|
69 <p><a name="index-fftw_005fmpi_005flocal_005fsize_005f2d-448"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005f3d-449"></a><a name="index-fftw_005fmpi_005flocal_005fsize-450"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005f2d_005ftransposed-451"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005f3d_005ftransposed-452"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005ftransposed-453"></a>
|
|
cannam@95
|
70 <pre class="example"> ptrdiff_t fftw_mpi_local_size_2d(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,
|
|
cannam@95
|
71 ptrdiff_t *local_n0, ptrdiff_t *local_0_start);
|
|
cannam@95
|
72 ptrdiff_t fftw_mpi_local_size_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@95
|
73 MPI_Comm comm,
|
|
cannam@95
|
74 ptrdiff_t *local_n0, ptrdiff_t *local_0_start);
|
|
cannam@95
|
75 ptrdiff_t fftw_mpi_local_size(int rnk, const ptrdiff_t *n, MPI_Comm comm,
|
|
cannam@95
|
76 ptrdiff_t *local_n0, ptrdiff_t *local_0_start);
|
|
cannam@95
|
77
|
|
cannam@95
|
78 ptrdiff_t fftw_mpi_local_size_2d_transposed(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm,
|
|
cannam@95
|
79 ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
|
|
cannam@95
|
80 ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
|
|
cannam@95
|
81 ptrdiff_t fftw_mpi_local_size_3d_transposed(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2,
|
|
cannam@95
|
82 MPI_Comm comm,
|
|
cannam@95
|
83 ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
|
|
cannam@95
|
84 ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
|
|
cannam@95
|
85 ptrdiff_t fftw_mpi_local_size_transposed(int rnk, const ptrdiff_t *n, MPI_Comm comm,
|
|
cannam@95
|
86 ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
|
|
cannam@95
|
87 ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
|
|
cannam@95
|
88 </pre>
|
|
cannam@95
|
89 <p>These functions return the number of elements to allocate (complex
|
|
cannam@95
|
90 numbers for DFT/r2c/c2r plans, real numbers for r2r plans), whereas
|
|
cannam@95
|
91 the <code>local_n0</code> and <code>local_0_start</code> return the portion
|
|
cannam@95
|
92 (<code>local_0_start</code> to <code>local_0_start + local_n0 - 1</code>) of the
|
|
cannam@95
|
93 first dimension of an n<sub>0</sub> × n<sub>1</sub> × n<sub>2</sub> × … × n<sub>d-1</sub> array that is stored on the local
|
|
cannam@95
|
94 process. See <a href="Basic-and-advanced-distribution-interfaces.html#Basic-and-advanced-distribution-interfaces">Basic and advanced distribution interfaces</a>. For
|
|
cannam@95
|
95 <code>FFTW_MPI_TRANSPOSED_OUT</code> plans, the ‘<samp><span class="samp">_transposed</span></samp>’ variants
|
|
cannam@95
|
96 are useful in order to also return the local portion of the first
|
|
cannam@95
|
97 dimension in the n<sub>1</sub> × n<sub>0</sub> × n<sub>2</sub> ×…× n<sub>d-1</sub> transposed output. See <a href="Transposed-distributions.html#Transposed-distributions">Transposed distributions</a>. The advanced interface for multidimensional
|
|
cannam@95
|
98 transforms is:
|
|
cannam@95
|
99
|
|
cannam@95
|
100 <p><a name="index-advanced-interface-454"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005fmany-455"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005fmany_005ftransposed-456"></a>
|
|
cannam@95
|
101 <pre class="example"> ptrdiff_t fftw_mpi_local_size_many(int rnk, const ptrdiff_t *n, ptrdiff_t howmany,
|
|
cannam@95
|
102 ptrdiff_t block0, MPI_Comm comm,
|
|
cannam@95
|
103 ptrdiff_t *local_n0, ptrdiff_t *local_0_start);
|
|
cannam@95
|
104 ptrdiff_t fftw_mpi_local_size_many_transposed(int rnk, const ptrdiff_t *n, ptrdiff_t howmany,
|
|
cannam@95
|
105 ptrdiff_t block0, ptrdiff_t block1, MPI_Comm comm,
|
|
cannam@95
|
106 ptrdiff_t *local_n0, ptrdiff_t *local_0_start,
|
|
cannam@95
|
107 ptrdiff_t *local_n1, ptrdiff_t *local_1_start);
|
|
cannam@95
|
108 </pre>
|
|
cannam@95
|
109 <p>These differ from the basic interface in only two ways. First, they
|
|
cannam@95
|
110 allow you to specify block sizes <code>block0</code> and <code>block1</code> (the
|
|
cannam@95
|
111 latter for the transposed output); you can pass
|
|
cannam@95
|
112 <code>FFTW_MPI_DEFAULT_BLOCK</code> to use FFTW's default block size as in
|
|
cannam@95
|
113 the basic interface. Second, you can pass a <code>howmany</code> parameter,
|
|
cannam@95
|
114 corresponding to the advanced planning interface below: this is for
|
|
cannam@95
|
115 transforms of contiguous <code>howmany</code>-tuples of numbers
|
|
cannam@95
|
116 (<code>howmany = 1</code> in the basic interface).
|
|
cannam@95
|
117
|
|
cannam@95
|
118 <p>The corresponding basic and advanced routines for one-dimensional
|
|
cannam@95
|
119 transforms (currently only complex DFTs) are:
|
|
cannam@95
|
120
|
|
cannam@95
|
121 <p><a name="index-fftw_005fmpi_005flocal_005fsize_005f1d-457"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005fmany_005f1d-458"></a>
|
|
cannam@95
|
122 <pre class="example"> ptrdiff_t fftw_mpi_local_size_1d(
|
|
cannam@95
|
123 ptrdiff_t n0, MPI_Comm comm, int sign, unsigned flags,
|
|
cannam@95
|
124 ptrdiff_t *local_ni, ptrdiff_t *local_i_start,
|
|
cannam@95
|
125 ptrdiff_t *local_no, ptrdiff_t *local_o_start);
|
|
cannam@95
|
126 ptrdiff_t fftw_mpi_local_size_many_1d(
|
|
cannam@95
|
127 ptrdiff_t n0, ptrdiff_t howmany,
|
|
cannam@95
|
128 MPI_Comm comm, int sign, unsigned flags,
|
|
cannam@95
|
129 ptrdiff_t *local_ni, ptrdiff_t *local_i_start,
|
|
cannam@95
|
130 ptrdiff_t *local_no, ptrdiff_t *local_o_start);
|
|
cannam@95
|
131 </pre>
|
|
cannam@95
|
132 <p><a name="index-FFTW_005fMPI_005fSCRAMBLED_005fOUT-459"></a><a name="index-FFTW_005fMPI_005fSCRAMBLED_005fIN-460"></a>As above, the return value is the number of elements to allocate
|
|
cannam@95
|
133 (complex numbers, for complex DFTs). The <code>local_ni</code> and
|
|
cannam@95
|
134 <code>local_i_start</code> arguments return the portion
|
|
cannam@95
|
135 (<code>local_i_start</code> to <code>local_i_start + local_ni - 1</code>) of the
|
|
cannam@95
|
136 1d array that is stored on this process for the transform
|
|
cannam@95
|
137 <em>input</em>, and <code>local_no</code> and <code>local_o_start</code> are the
|
|
cannam@95
|
138 corresponding quantities for the input. The <code>sign</code>
|
|
cannam@95
|
139 (<code>FFTW_FORWARD</code> or <code>FFTW_BACKWARD</code>) and <code>flags</code> must
|
|
cannam@95
|
140 match the arguments passed when creating a plan. Although the inputs
|
|
cannam@95
|
141 and outputs have different data distributions in general, it is
|
|
cannam@95
|
142 guaranteed that the <em>output</em> data distribution of an
|
|
cannam@95
|
143 <code>FFTW_FORWARD</code> plan will match the <em>input</em> data distribution
|
|
cannam@95
|
144 of an <code>FFTW_BACKWARD</code> plan and vice versa; similarly for the
|
|
cannam@95
|
145 <code>FFTW_MPI_SCRAMBLED_OUT</code> and <code>FFTW_MPI_SCRAMBLED_IN</code> flags.
|
|
cannam@95
|
146 See <a href="One_002ddimensional-distributions.html#One_002ddimensional-distributions">One-dimensional distributions</a>.
|
|
cannam@95
|
147
|
|
cannam@95
|
148 </body></html>
|
|
cannam@95
|
149
|
