Mercurial > hg > sv-dependency-builds
view src/fftw-3.3.3/doc/html/MPI-Data-Distribution-Functions.html @ 10:37bf6b4a2645
Add FFTW3
| author | Chris Cannam |
|---|---|
| date | Wed, 20 Mar 2013 15:35:50 +0000 |
| parents | |
| children |
line wrap: on
line source
<html lang="en"> <head> <title>MPI Data Distribution Functions - FFTW 3.3.3</title> <meta http-equiv="Content-Type" content="text/html"> <meta name="description" content="FFTW 3.3.3"> <meta name="generator" content="makeinfo 4.13"> <link title="Top" rel="start" href="index.html#Top"> <link rel="up" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference" title="FFTW MPI Reference"> <link rel="prev" href="Using-MPI-Plans.html#Using-MPI-Plans" title="Using MPI Plans"> <link rel="next" href="MPI-Plan-Creation.html#MPI-Plan-Creation" title="MPI Plan Creation"> <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage"> <!-- This manual is for FFTW (version 3.3.3, 25 November 2012). Copyright (C) 2003 Matteo Frigo. Copyright (C) 2003 Massachusetts Institute of Technology. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that this permission notice may be stated in a translation approved by the Free Software Foundation. --> <meta http-equiv="Content-Style-Type" content="text/css"> <style type="text/css"><!-- pre.display { font-family:inherit } pre.format { font-family:inherit } pre.smalldisplay { font-family:inherit; font-size:smaller } pre.smallformat { font-family:inherit; font-size:smaller } pre.smallexample { font-size:smaller } pre.smalllisp { font-size:smaller } span.sc { font-variant:small-caps } span.roman { font-family:serif; font-weight:normal; } span.sansserif { font-family:sans-serif; font-weight:normal; } --></style> </head> <body> <div class="node"> <a name="MPI-Data-Distribution-Functions"></a> <p> Next: <a rel="next" accesskey="n" href="MPI-Plan-Creation.html#MPI-Plan-Creation">MPI Plan Creation</a>, Previous: <a rel="previous" accesskey="p" href="Using-MPI-Plans.html#Using-MPI-Plans">Using MPI Plans</a>, Up: <a rel="up" accesskey="u" href="FFTW-MPI-Reference.html#FFTW-MPI-Reference">FFTW MPI Reference</a> <hr> </div> <h4 class="subsection">6.12.4 MPI Data Distribution Functions</h4> <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 allocate your arrays, <em>before</em> creating a plan, you must first call one of the following routines to determine the required allocation size and the portion of the array locally stored on a given process. The <code>MPI_Comm</code> communicator passed here must be equivalent to the communicator used below for plan creation. <p>The basic interface for multidimensional transforms consists of the functions: <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> <pre class="example"> ptrdiff_t fftw_mpi_local_size_2d(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start); ptrdiff_t fftw_mpi_local_size_3d(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start); ptrdiff_t fftw_mpi_local_size(int rnk, const ptrdiff_t *n, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start); ptrdiff_t fftw_mpi_local_size_2d_transposed(ptrdiff_t n0, ptrdiff_t n1, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start, ptrdiff_t *local_n1, ptrdiff_t *local_1_start); ptrdiff_t fftw_mpi_local_size_3d_transposed(ptrdiff_t n0, ptrdiff_t n1, ptrdiff_t n2, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start, ptrdiff_t *local_n1, ptrdiff_t *local_1_start); ptrdiff_t fftw_mpi_local_size_transposed(int rnk, const ptrdiff_t *n, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start, ptrdiff_t *local_n1, ptrdiff_t *local_1_start); </pre> <p>These functions return the number of elements to allocate (complex numbers for DFT/r2c/c2r plans, real numbers for r2r plans), whereas the <code>local_n0</code> and <code>local_0_start</code> return the portion (<code>local_0_start</code> to <code>local_0_start + local_n0 - 1</code>) of the 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 process. See <a href="Basic-and-advanced-distribution-interfaces.html#Basic-and-advanced-distribution-interfaces">Basic and advanced distribution interfaces</a>. For <code>FFTW_MPI_TRANSPOSED_OUT</code> plans, the ‘<samp><span class="samp">_transposed</span></samp>’ variants are useful in order to also return the local portion of the first 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 transforms is: <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> <pre class="example"> ptrdiff_t fftw_mpi_local_size_many(int rnk, const ptrdiff_t *n, ptrdiff_t howmany, ptrdiff_t block0, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start); ptrdiff_t fftw_mpi_local_size_many_transposed(int rnk, const ptrdiff_t *n, ptrdiff_t howmany, ptrdiff_t block0, ptrdiff_t block1, MPI_Comm comm, ptrdiff_t *local_n0, ptrdiff_t *local_0_start, ptrdiff_t *local_n1, ptrdiff_t *local_1_start); </pre> <p>These differ from the basic interface in only two ways. First, they allow you to specify block sizes <code>block0</code> and <code>block1</code> (the latter for the transposed output); you can pass <code>FFTW_MPI_DEFAULT_BLOCK</code> to use FFTW's default block size as in the basic interface. Second, you can pass a <code>howmany</code> parameter, corresponding to the advanced planning interface below: this is for transforms of contiguous <code>howmany</code>-tuples of numbers (<code>howmany = 1</code> in the basic interface). <p>The corresponding basic and advanced routines for one-dimensional transforms (currently only complex DFTs) are: <p><a name="index-fftw_005fmpi_005flocal_005fsize_005f1d-457"></a><a name="index-fftw_005fmpi_005flocal_005fsize_005fmany_005f1d-458"></a> <pre class="example"> ptrdiff_t fftw_mpi_local_size_1d( ptrdiff_t n0, MPI_Comm comm, int sign, unsigned flags, ptrdiff_t *local_ni, ptrdiff_t *local_i_start, ptrdiff_t *local_no, ptrdiff_t *local_o_start); ptrdiff_t fftw_mpi_local_size_many_1d( ptrdiff_t n0, ptrdiff_t howmany, MPI_Comm comm, int sign, unsigned flags, ptrdiff_t *local_ni, ptrdiff_t *local_i_start, ptrdiff_t *local_no, ptrdiff_t *local_o_start); </pre> <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 (complex numbers, for complex DFTs). The <code>local_ni</code> and <code>local_i_start</code> arguments return the portion (<code>local_i_start</code> to <code>local_i_start + local_ni - 1</code>) of the 1d array that is stored on this process for the transform <em>input</em>, and <code>local_no</code> and <code>local_o_start</code> are the corresponding quantities for the input. The <code>sign</code> (<code>FFTW_FORWARD</code> or <code>FFTW_BACKWARD</code>) and <code>flags</code> must match the arguments passed when creating a plan. Although the inputs and outputs have different data distributions in general, it is guaranteed that the <em>output</em> data distribution of an <code>FFTW_FORWARD</code> plan will match the <em>input</em> data distribution of an <code>FFTW_BACKWARD</code> plan and vice versa; similarly for the <code>FFTW_MPI_SCRAMBLED_OUT</code> and <code>FFTW_MPI_SCRAMBLED_IN</code> flags. See <a href="One_002ddimensional-distributions.html#One_002ddimensional-distributions">One-dimensional distributions</a>. </body></html>