|
d@0
|
1 <html lang="en">
|
|
d@0
|
2 <head>
|
|
d@0
|
3 <title>Distributed-memory FFTW with MPI - FFTW 3.2alpha3</title>
|
|
d@0
|
4 <meta http-equiv="Content-Type" content="text/html">
|
|
d@0
|
5 <meta name="description" content="FFTW 3.2alpha3">
|
|
d@0
|
6 <meta name="generator" content="makeinfo 4.8">
|
|
d@0
|
7 <link title="Top" rel="start" href="index.html#Top">
|
|
d@0
|
8 <link rel="prev" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW" title="Multi-threaded FFTW">
|
|
d@0
|
9 <link rel="next" href="FFTW-on-the-Cell-Processor.html#FFTW-on-the-Cell-Processor" title="FFTW on the Cell Processor">
|
|
d@0
|
10 <link href="http://www.gnu.org/software/texinfo/" rel="generator-home" title="Texinfo Homepage">
|
|
d@0
|
11 <!--
|
|
d@0
|
12 This manual is for FFTW
|
|
d@0
|
13 (version 3.2alpha3, 14 August 2007).
|
|
d@0
|
14
|
|
d@0
|
15 Copyright (C) 2003 Matteo Frigo.
|
|
d@0
|
16
|
|
d@0
|
17 Copyright (C) 2003 Massachusetts Institute of Technology.
|
|
d@0
|
18
|
|
d@0
|
19 Permission is granted to make and distribute verbatim copies of
|
|
d@0
|
20 this manual provided the copyright notice and this permission
|
|
d@0
|
21 notice are preserved on all copies.
|
|
d@0
|
22
|
|
d@0
|
23 Permission is granted to copy and distribute modified versions of
|
|
d@0
|
24 this manual under the conditions for verbatim copying, provided
|
|
d@0
|
25 that the entire resulting derived work is distributed under the
|
|
d@0
|
26 terms of a permission notice identical to this one.
|
|
d@0
|
27
|
|
d@0
|
28 Permission is granted to copy and distribute translations of this
|
|
d@0
|
29 manual into another language, under the above conditions for
|
|
d@0
|
30 modified versions, except that this permission notice may be
|
|
d@0
|
31 stated in a translation approved by the Free Software Foundation.
|
|
d@0
|
32 -->
|
|
d@0
|
33 <meta http-equiv="Content-Style-Type" content="text/css">
|
|
d@0
|
34 <style type="text/css"><!--
|
|
d@0
|
35 pre.display { font-family:inherit }
|
|
d@0
|
36 pre.format { font-family:inherit }
|
|
d@0
|
37 pre.smalldisplay { font-family:inherit; font-size:smaller }
|
|
d@0
|
38 pre.smallformat { font-family:inherit; font-size:smaller }
|
|
d@0
|
39 pre.smallexample { font-size:smaller }
|
|
d@0
|
40 pre.smalllisp { font-size:smaller }
|
|
d@0
|
41 span.sc { font-variant:small-caps }
|
|
d@0
|
42 span.roman { font-family:serif; font-weight:normal; }
|
|
d@0
|
43 span.sansserif { font-family:sans-serif; font-weight:normal; }
|
|
d@0
|
44 --></style>
|
|
d@0
|
45 </head>
|
|
d@0
|
46 <body>
|
|
d@0
|
47 <div class="node">
|
|
d@0
|
48 <p>
|
|
d@0
|
49 <a name="Distributed-memory-FFTW-with-MPI"></a>
|
|
d@0
|
50 <a name="Distributed_002dmemory-FFTW-with-MPI"></a>
|
|
d@0
|
51 Next: <a rel="next" accesskey="n" href="FFTW-on-the-Cell-Processor.html#FFTW-on-the-Cell-Processor">FFTW on the Cell Processor</a>,
|
|
d@0
|
52 Previous: <a rel="previous" accesskey="p" href="Multi_002dthreaded-FFTW.html#Multi_002dthreaded-FFTW">Multi-threaded FFTW</a>,
|
|
d@0
|
53 Up: <a rel="up" accesskey="u" href="index.html#Top">Top</a>
|
|
d@0
|
54 <hr>
|
|
d@0
|
55 </div>
|
|
d@0
|
56
|
|
d@0
|
57 <h2 class="chapter">6 Distributed-memory FFTW with MPI</h2>
|
|
d@0
|
58
|
|
d@0
|
59 <p><a name="index-MPI-329"></a>
|
|
d@0
|
60 <a name="index-parallel-transform-330"></a>In this chapter we document the parallel FFTW routines for parallel
|
|
d@0
|
61 hardware supporting the MPI message-passing interface. Unlike the
|
|
d@0
|
62 shared-memory threads described in the previous chapter, MPI allows
|
|
d@0
|
63 you to use <em>distributed-memory</em> parallelism, where each CPU has
|
|
d@0
|
64 its own separate memory, and which can scale up to clusters of many
|
|
d@0
|
65 thousands of processors. This capability comes at a price, however:
|
|
d@0
|
66 each process only stores a <em>portion</em> of the data to be
|
|
d@0
|
67 transformed, which means that the data structures and
|
|
d@0
|
68 programming-interface are quite different from the serial or threads
|
|
d@0
|
69 versions of FFTW.
|
|
d@0
|
70 <a name="index-data-distribution-331"></a>
|
|
d@0
|
71 Distributed-memory parallelism is especially useful when you are
|
|
d@0
|
72 transforming arrays so large that they do not fit into the memory of a
|
|
d@0
|
73 single processor. The storage per-process required by FFTW's MPI
|
|
d@0
|
74 routines is proportional to the total array size divided by the number
|
|
d@0
|
75 of processes. Conversely, distributed-memory parallelism can easily
|
|
d@0
|
76 pose an unacceptably high communications overhead for small problems;
|
|
d@0
|
77 the threshold problem size for which parallelism becomes important
|
|
d@0
|
78 will depend on the precise problem you are interested in, your
|
|
d@0
|
79 hardware, and your MPI implementation.
|
|
d@0
|
80
|
|
d@0
|
81 <p>A note on terminology: in MPI, you divide the data among a set of
|
|
d@0
|
82 “processes” which each run in their own memory address space.
|
|
d@0
|
83 Generally, each process runs on a different physical processor, but
|
|
d@0
|
84 this is not required. A group of processes in MPI is described by an
|
|
d@0
|
85 opaque data structure called a “communicator,” the most common of
|
|
d@0
|
86 which is the predefined communicator <code>MPI_COMM_WORLD</code> which
|
|
d@0
|
87 refers to <em>all</em> processes. For more information on these and
|
|
d@0
|
88 other concepts common to all MPI programs, we refer the reader to the
|
|
d@0
|
89 documentation at <a href="http://www-unix.mcs.anl.gov/mpi/">the MPI home page</a>.
|
|
d@0
|
90 <a name="index-MPI-communicator-332"></a><a name="index-MPI_005fCOMM_005fWORLD-333"></a>
|
|
d@0
|
91 We assume in this chapter that the reader is familiar with the usage
|
|
d@0
|
92 of the serial (uniprocessor) FFTW, and focus only on the concepts new
|
|
d@0
|
93 to the MPI interface.
|
|
d@0
|
94
|
|
d@0
|
95 <ul class="menu">
|
|
d@0
|
96 <li><a accesskey="1" href="FFTW-MPI-Installation.html#FFTW-MPI-Installation">FFTW MPI Installation</a>
|
|
d@0
|
97 <li><a accesskey="2" href="Linking-and-Initializing-MPI-FFTW.html#Linking-and-Initializing-MPI-FFTW">Linking and Initializing MPI FFTW</a>
|
|
d@0
|
98 <li><a accesskey="3" href="Simple-MPI-example.html#Simple-MPI-example">Simple MPI example</a>
|
|
d@0
|
99 <li><a accesskey="4" href="MPI-data-distribution.html#MPI-data-distribution">MPI data distribution</a>
|
|
d@0
|
100 <li><a accesskey="5" href="Multi_002ddimensional-MPI-DFT-of-Real-Data.html#Multi_002ddimensional-MPI-DFT-of-Real-Data">Multi-dimensional MPI DFT of Real Data</a>
|
|
d@0
|
101 <li><a accesskey="6" href="Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms.html#Other-Multi_002ddimensional-Real_002ddata-MPI-Transforms">Other Multi-dimensional Real-data MPI Transforms</a>
|
|
d@0
|
102 <li><a accesskey="7" href="FFTW-MPI-Transposes.html#FFTW-MPI-Transposes">FFTW MPI Transposes</a>
|
|
d@0
|
103 <li><a accesskey="8" href="FFTW-MPI-Wisdom.html#FFTW-MPI-Wisdom">FFTW MPI Wisdom</a>
|
|
d@0
|
104 <li><a accesskey="9" href="Avoiding-MPI-Deadlocks.html#Avoiding-MPI-Deadlocks">Avoiding MPI Deadlocks</a>
|
|
d@0
|
105 <li><a href="FFTW-MPI-Performance-Tips.html#FFTW-MPI-Performance-Tips">FFTW MPI Performance Tips</a>
|
|
d@0
|
106 <li><a href="Combining-MPI-and-Threads.html#Combining-MPI-and-Threads">Combining MPI and Threads</a>
|
|
d@0
|
107 <li><a href="FFTW-MPI-Reference.html#FFTW-MPI-Reference">FFTW MPI Reference</a>
|
|
d@0
|
108 </ul>
|
|
d@0
|
109
|
|
d@0
|
110 <!-- -->
|
|
d@0
|
111 </body></html>
|
|
d@0
|
112
|
