forked from xiaolongma/jburkardt-f
-
Notifications
You must be signed in to change notification settings - Fork 1
/
openmp.html
552 lines (493 loc) · 17.3 KB
/
openmp.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
<html>
<head>
<title>
OPENMP - FORTRAN90 Examples of Parallel Programming with OpenMP
</title>
</head>
<body bgcolor="#EEEEEE" link="#CC0000" alink="#FF3300" vlink="#000055">
<h1 align = "center">
OPENMP <br> FORTRAN90 Examples of Parallel Programming with OpenMP
</h1>
<hr>
<p>
<b>OPENMP</b>
is a directory of FORTRAN90 examples which
illustrate the use of the OpenMP application program interface
for carrying out parallel computations in a shared memory environment.
</p>
<p>
The directives
allow the user to mark areas of the code, such as <b>do</b>,
<b>while</b> or <b>for</b> loops, which are suitable for parallel
processing. The directives appear as a special kind of
comment, so the program can be compiled and run in serial mode.
However, the user can tell the compiler to "notice" the special
directives, in which case a version of the program will be created
that runs in parallel.
</p>
<p>
Thus the same program can easily be run in serial or parallel mode
on a given computer, or run on a computer that does not have
<b>OpenMP</b> at all.
</p>
<p>
<b>OpenMP</b> is suitable for a <i>shared memory</i> parallel system,
that is, a situation in which there is a single memory space, and
multiple processors. If memory is shared, then typically the number
of processors will be small, and they will all be on the same
physical machine.
</p>
<p>
By contrast, in a <i>distributed memory</i> system, items of data
are closely associated with a particular processor. There may
be a very large number of processors, and they may be more loosely
coupled and even on different machines. Such a system will need
to be handled with <a href = "../mpi/mpi.html">MPI</a> or some
other message passing interface.
</p>
<p>
<b>OpenMP</b> descended in part from the old Cray microtasking
directives, so if you've lived long enough to remember those, you
will recognize some features.
</p>
<h3 align = "center">
FORTRAN90 Issues
</h3>
<p>
OpenMP includes a number of functions whose type must be declared in
any program that uses them. To avoid having to declare these functions,
you can use the command
<pre><b>
use omp_lib
</b></pre>
in any routine that invokes OpenMP functions.
</p>
<h3 align = "center">
"Pretend" Parallelism
</h3>
<p>
<b>OpenMP</b> allows you to "request" any number of threads of execution.
This is a request, and it's not always a wise request. If your system has
four processors available, and they're not busy doing other things, or serving
other users, maybe 4 threads is what you want. But you can't guarantee
you'll get the undivided use of those processors. Moreover, if you run the
same program using 1 thread and 4 threads, you may find that using 4 threads
<i>slows you down</i>, either because you don't actually have 4 processors,
(so the system has the overhead of pretending to run in parallel), or because
the processors you have are also busy doing other things.
</p>
<p>
For this reason, it's wise to run the program at least once in single thread mode,
so you have a benchmark against which to measure the speedup you got (or didn't
get!) versus the speedup you hoped for.
</p>
<h3 align = "center">
Compiler Support
</h3>
<p>
The compiler you use must recognize the OpenMP directives in order to produce
code that will run in parallel. Here are some of the compilers available that
support OpenMP:
<ul>
<li>
The GNU compilers <b>gcc/g++</b> (version 4.2 or later)
and <b>gfortran</b> (version 2.0 or later).
</li>
<li>
The IBM compilers <b>xlc</b> and <b>xlf</b>.
</li>
<li>
The Intel compilers <b>icc</b>, <b>icpc</b> and <b>ifort</b>.
</li>
<li>
The Microsoft Visual C++ compiler.
</li>
<li>
The Portland Group C/C++ and Fortran compilers.
</li>
<li>
Sun Studio C/C++ and Fortran95 compilers.
</li>
</ul>
</p>
<h3 align = "center">
Licensing:
</h3>
<p>
The computer code and data files described and made available on this web page
are distributed under
<a href = "../../txt/gnu_lgpl.txt">the GNU LGPL license.</a>
</p>
<h3 align = "center">
Languages:
</h3>
<p>
<b>OPENMP</b> examples are available in
<a href = "../../c_src/openmp/openmp.html">a C version</a> and
<a href = "../../cpp_src/openmp/openmp.html">a C++ version</a> and
<a href = "../../f77_src/openmp/openmp.html">a FORTRAN77 version</a> and
<a href = "../../f_src/openmp/openmp.html">a FORTRAN90 version</a>.
</p>
<h3 align = "center">
Related Data and Programs:
</h3>
<p>
<a href = "../../f_src/dijkstra_openmp/dijkstra_openmp.html">
DIJKSTRA_OPENMP</a>,
a FORTRAN90 program which
uses OpenMP to parallelize a simple example of Dijkstra's
minimum distance algorithm for graphs.
</p>
<p>
<a href = "../../f_src/fft_openmp/fft_openmp.html">
FFT_OPENMP</a>,
a FORTRAN90 program which
demonstrates the computation of a Fast Fourier Transform
in parallel, using OpenMP.
</p>
<p>
<a href = "../../f_src/heated_plate_openmp/heated_plate_openmp.html">
HEATED_PLATE_OPENMP</a>,
a FORTRAN90 program which
solves the steady (time independent) heat equation in a 2D
rectangular region, using OpenMP to run in parallel.
</p>
<p>
<a href = "../../f_src/hello_openmp/hello_openmp.html">
HELLO_OPENMP</a>,
a FORTRAN90 program which
prints out "Hello, world!" using the OpenMP parallel programming environment.
</p>
<p>
<a href = "../../f_src/mandelbrot_openmp/mandelbrot_openmp.html">
MANDELBROT_OPENMP</a>,
a FORTRAN90 program which
generates an ASCII Portable Pixel Map (PPM) image of the Mandelbrot fractal set,
using OpenMP for parallel execution.
</p>
<p>
<a href = "../../f_src/md_openmp/md_openmp.html">
MD_OPENMP</a>,
a FORTRAN90 program which
carries out a molecular dynamics simulation using OpenMP.
<p>
<p>
<a href = "../../f_src/mpi/mpi.html">
MPI</a>,
FORTRAN90 programs which
illustrate the use of parallel programming in a distributed memory
environment, using message passing.
</p>
<p>
<a href = "../../f_src/multitask_openmp/multitask_openmp.html">
MULTITASK_OPENMP</a>,
a FORTRAN90 program which
demonstrates how to "multitask", that is, to execute several unrelated
and distinct tasks simultaneously, using OpenMP for parallel execution.
</p>
<p>
<a href = "../../f_src/mxv_openmp/mxv_openmp.html">
MXM_OPENMP</a>,
a FORTRAN90 program which
computes a dense matrix product C=A*B,
using OpenMP for parallel execution.
<p>
<p>
<a href = "../../f_src/mxv_openmp/mxv_openmp.html">
MXV_OPENMP</a>,
a FORTRAN90 program which
compares the performance of plain vanilla Fortran and the FORTRAN90
intrinsic routine MATMUL, for the matrix multiplication problem
y=A*x, with and without parallelization by OpenMP.
<p>
<p>
<a href = "../../f_src/openmp_stubs/openmp_stubs.html">
OPENMP_STUBS</a>,
a FORTRAN90 library which
implements a "stub" version of OpenMP, so that an OpenMP program
can be compiled, linked and executed on a system that does not
have OpenMP installed.
</p>
<p>
<a href = "../../f_src/poisson_openmp/poisson_openmp.html">
POISSON_OPENMP</a>,
a FORTRAN90 program which
computes an approximate solution to the Poisson equation in a rectangle,
using the Jacobi iteration to solve the linear system, and OpenMP to
carry out the Jacobi iteration in parallel.
</p>
<p>
<a href = "../../f_src/prime_openmp/prime_openmp.html">
PRIME_OPENMP</a>,
a FORTRAN90 program which
counts the number of primes between 1 and N, using OpenMP for parallel execution.
</p>
<p>
<a href = "../../c_src/pthreads/pthreads.html">
PTHREADS</a>,
C programs which
illustrate the use of the POSIX thread library to carry out
parallel program execution.
</p>
<p>
<a href = "../../f_src/quad_openmp/quad_openmp.html">
QUAD_OPENMP</a>,
a FORTRAN90 program which
approximates an integral using a quadrature rule, and carries out the
computation in parallel using OpenMP.
</p>
<p>
<a href = "../../f_src/random_openmp/random_openmp.html">
RANDOM_OPENMP</a>,
a FORTRAN90 program which
illustrates how a parallel program using OpenMP can generate multiple
distinct streams of random numbers.
</p>
<p>
<a href = "../../f_src/satisfy_openmp/satisfy_openmp.html">
SATISFY_OPENMP</a>,
a FORTRAN90 program which
demonstrates, for a particular circuit, an exhaustive search
for solutions of the circuit satisfiability problem,
using OpenMP for parallel execution.
<p>
<p>
<a href = "../../f_src/schedule_openmp/schedule_openmp.html">
SCHEDULE_OPENMP</a>,
a FORTRAN90 program which
demonstrates the default, static, and dynamic methods of "scheduling"
loop iterations in OpenMP to avoid work imbalance.
<p>
<p>
<a href = "../../f_src/sgefa_openmp/sgefa_openmp.html">
SGEFA_OPENMP</a>,
a FORTRAN90 program which
reimplements the SGEFA/SGESL linear algebra routines from
LINPACK for use with OpenMP.
</p>
<p>
<a href = "../../f_src/ziggurat_openmp/ziggurat_openmp.html">
ZIGGURAT_OPENMP</a>,
a FORTRAN90 program which
demonstrates how the ZIGGURAT library can be used to generate random numbers
in an OpenMP parallel program.
</p>
<h3 align = "center">
Reference:
</h3>
<p>
<ol>
<li>
Peter Arbenz, Wesley Petersen,<br>
Introduction to Parallel Computing - A practical guide with examples in C,<br>
Oxford University Press,<br>
ISBN: 0-19-851576-6,<br>
LC: QA76.58.P47.
</li>
<li>
Rohit Chandra, Leonardo Dagum, Dave Kohr, Dror Maydan,
Jeff McDonald, Ramesh Menon,<br>
Parallel Programming in OpenMP,<br>
Morgan Kaufmann, 2001,<br>
ISBN: 1-55860-671-8,<br>
LC: QA76.642.P32.
</li>
<li>
Barbara Chapman, Gabriele Jost, Ruud vanderPas, David Kuck,<br>
Using OpenMP: Portable Shared Memory Parallel Processing,<br>
MIT Press, 2007,<br>
ISBN13: 978-0262533027,<br>
LC: QA76.642.C49.
</li>
<li>
<a href = "http://www.openmp.org/">The OpenMP web site</a>
</li>
<li>
OpenMP Architecture Review Board,<br>
OpenMP Application Program Interface,<br>
Version 3.0,<br>
May 2008.
</li>
</ol>
</p>
<h3 align = "center">
Examples and Tests:
</h3>
<p>
<b>COMPUTE_PI</b> shows how information can be shared.
Several processors need to compute pieces of a sum that will
approximate pi.
<ul>
<li>
<a href = "compute_pi.f90">compute_pi.f90</a>,
the source code;
</li>
<li>
<a href = "compute_pi_ithaca.sh">compute_pi_ithaca.sh</a>,
BASH commands to compile, link, load and run the program
on Virginia Tech's Ithaca system.
</li>
<li>
<a href = "compute_pi_local.sh">compute_pi_local.sh</a>,
BASH commands to compile, link, load and run the program
locally.
</li>
<li>
<a href = "compute_pi_local_output.txt">compute_pi_local_output.txt</a>,
the output file;
</li>
</ul>
</p>
<p>
<b>DOT_PRODUCT</b> compares the computation of a vector dot product
in sequential mode, and using OpenMP. Typically, the overhead of using
parallel processing outweighs the advantage for small vector sizes N.
The code demonstrates this fact by using a number of values of N,
and by running both sequential and OpenMP versions of the calculation.
<ul>
<li>
<a href = "dot_product.f90">dot_product.f90</a>,
the source code;
</li>
<li>
<a href = "dot_product_local.sh">dot_product_local.sh</a>,
BASH commands to compile, link, load and run the program.
</li>
<li>
<a href = "dot_product_local_output.txt">dot_product_local_output.txt</a>,
the output file;
</li>
</ul>
</p>
<p>
<b>HELMHOLTZ</b> is a more extensive program that solves
the Helmholtz equation on a regular grid, using a Jacobi iterative
linear equation solver with overrelaxation;
<ul>
<li>
<a href = "helmholtz.f90">helmholtz.f90</a>,
the source code;
</li>
<li>
<a href = "helmholtz_local.sh">helmholtz_local.sh</a>,
BASH commands to compile, link, load and run the program
using 4 threads.
</li>
<li>
<a href = "helmholtz_local_output.txt">helmholtz_local_output.txt</a>,
the output file;
</li>
</ul>
</p>
<p>
<b>MAXIMUM</b> shows how FORTRAN programs are allowed to compute
the maximum entry of a vector as a reduction variable.
<ul>
<li>
<a href = "maximum.f90">maximum.f90</a>,
the source code;
</li>
<li>
<a href = "maximum_local.sh">maximum_local.sh</a>,
BASH commands to compile, link, load and run the program.
</li>
<li>
<a href = "maximum_local_output.txt">maximum_local_output.txt</a>,
the output file;
</li>
</ul>
</p>
<p>
<b>MXM</b> is a simple exercise in timing the computation of a matrix-matrix product.
<ul>
<li>
<a href = "mxm.f90">mxm.f90</a>,
the source code;
</li>
<li>
<a href = "mxm_local.sh">mxm_local.sh</a>,
BASH commands to compile, link, load and run the program
using 1 and 4 threads.
</li>
<li>
<a href = "mxm_local_output.txt">mxm_local_output.txt</a>,
the output file using 1 thread;
</li>
</ul>
</p>
<p>
<b>MXM2</b> repeats the MXM example, but tries to measure the CPU time
taken by individual threads. Unfortunately, the FORTRAN90 CPU_TIME function
is not required, by the language standard, to store and return separate
times for separate threads. So, whether this example gives you a good timing
or not depends on what compiler you use and where.
<ul>
<li>
<a href = "mxm2.f90">mxm2.f90</a>,
the source code;
</li>
<li>
<a href = "mxm2_local.sh">mxm2_local.sh</a>,
BASH commands to compile, link, load and run the program
using 1 or 2 threads.
</li>
<li>
<a href = "mxm2_local_output.txt">mxm2_local_output.txt</a>,
the output file using 1 threads;
</li>
</ul>
</p>
<p>
<b>RANDOM_NUMBER_CONTENTION</b> is a program which demontrates the possibility
that a program calling <b>random_number</b> can run MORE SLOWLY when parallel
threads are added. The performance degradation is presumed to be caused by
contention among the threads for access to the internal saved variables that
control the state of the random number generator.
<ul>
<li>
<a href = "random_number_contention.f90">random_number_contention.f90</a>,
the source code.
</li>
<li>
<a href = "random_number_contention.sh">random_number_contention.sh</a>,
commands to compile, link and run the source code.
</li>
<li>
<a href = "random_number_contention_output.txt">
random_number_contention_output.txt</a>,
the output file.
</li>
</ul>
</p>
<p>
<b>RANDOM_SEED</b> is a program which explores the possibility of
recreating exactly a stream of random numbers that were computed sequentially
by carefully copying all the input seed values, and then regenerating
the sequence in parallel.
<ul>
<li>
<a href = "random_seed.f90">random_seed.f90</a>,
the source code.
</li>
<li>
<a href = "random_seed.sh">random_seed.sh</a>,
commands to compile, link and run the source code.
</li>
<li>
<a href = "random_seed_output.txt">
random_seed.txt</a>,
the output file.
</li>
</ul>
</p>
<p>
You can go up one level to <a href = "../f_src.html">
the FORTRAN90 source codes</a>.
</p>
<hr>
<i>
Last revised on 03 September 2012.
</i>
<!-- John Burkardt -->
</body>
</html>