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
|
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE html
PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>Carpet — AMR for Cactus</title>
</head>
<body>
<h1 align="center">Carpet — Adaptive Mesh Refinement for the
Cactus Framework</h1>
<table>
<tr>
<td valign="top" bgcolor="#ffcc77" width="1%">
<a href="logo/Sierpinski.pdf"><img src="logo/Sierpinski.png" width="150" alt="Carpet logo (a Sierpiński carpet)" /></a>
<p><b>CarpetCode</b><br />
<a href="http://www.carpetcode.org/">home page</a></p>
<p><b>Documentation</b><br />
<a href="doc/documentation.pdf">Introduction</a> (PDF, 170 kB)<br />
<a href="doc/first-steps.pdf">First Steps</a> (PDF, 90 kB)<br />
<a href="doc/scheduling.pdf">Scheduling</a> (PDF, 160 kB)<br />
<a href="doc/domain-decomposition.pdf">Grid Structure</a> (PDF, 120 kB)<br />
<a href="doc/internals.pdf">Internals</a> (PDF, 130 kB)<br />
<a href="humour.html">Other Carpets</a></p>
<p><b>Mailing Lists</b><br />
<a href="http://lists.carpetcode.org/listinfo/developers/">Subscribe</a><br />
<a href="http://lists.carpetcode.org/archives/developers/">List Archive</a><br />
<a href="http://lists.carpetcode.org/listinfo/carpet-cvs/">CVS messages</a><br />
<a href="http://lists.carpetcode.org/listinfo/carpet-darcs/">darcs/git messages</a></p>
<p><b>Development</b><br />
<a href="get-carpet.html">Download</a><br />
<a href="http://bugs.carpetcode.org/">Bug Reports</a><br />
<a href="contributors.html">Contributors</a></p>
<!-- These are outdated
<a href="feature-requests.html">Missing features</a><br />
<a href="status-reports.html">Status reports</a><br />
-->
<p><b>Visualisation</b><br />
<a href="visualisation-tools.html">Tools</a><br />
<a href="https://mailserv.aei.mpg.de/mailman/listinfo/visualization/">Mailing List</a></p>
<p><b>Results</b><br />
<a href="publications/publications.html">Publications</a></p>
<p><b>Related</b><br />
<a href="http://www.cactuscode.org/">Cactus</a><br />
<a href="http://relativity.phys.lsu.edu/">LSU Relativity Group</a><br />
<a href="http://www.cct.lsu.edu/numerical/">numrel@CCT</a><br />
<a href="http://numrel.aei.mpg.de/">numrel@aei</a><br />
<a href="http://www.whiskycode.org/">Whisky</a><br />
<a href="http://relativity.phys.lsu.edu/postdocs/matt/taka.php">Taka</a><br />
<a href="http://sbir.nasa.gov/SBIR/abstracts/05/sttr/phase1/STTR-05-1-T4.02-9864.html?solicitationId=STTR_05_P1">ParCa</a></p>
<p><b>Carpet Users</b><br />
<a href="http://numrel.aei.mpg.de/">AEI Potsdam</a><br />
<!-- <a href="http://www.as.arizona.edu/">University of Arizona</a><br /> -->
<!-- <a href="http://www.astro.auth.gr/Science-Subjects/Gravity.html">AUTH</a><br /> -->
<a href="http://www.tapir.caltech.edu/">Caltech</a><br />
<!-- <a href="http://astrosun2.astro.cornell.edu/research/projects/blackholes/">Cornell</a><br /> -->
<a href="http://www.cra.gatech.edu/">Georgia Tech</a><br />
<a href="http://ww2.tpi.uni-jena.de/gravity/">Jena</a><br />
<a href="http://www.kisti.re.kr/english/">KISTI</a><br />
<a href="http://www.cct.lsu.edu/">LSU</a><br />
<!-- <a href="http://www.mpa-garching.mpg.de/hydro/index.shtml">MPA Garching</a><br /> -->
<a href="http://astrophysics.gsfc.nasa.gov/gravity/research/numrel.html">NASA Goddard</a><br />
<a href="http://www.fis.unipr.it/numrel/">Parma</a><br />
<a href="http://www.gravity.psu.edu/numrel/">Penn State</a><br />
<a href="http://astrophysics.rit.edu/">RIT</a><br />
<a href="http://www.sissa.it/ap/RelAstro/">SISSA</a><br />
<a href="http://www.maths.soton.ac.uk/applied/relativity/">Southampton</a><br />
<a href="http://www.tat.physik.uni-tuebingen.de/">TAT/CPT</a><br />
<a href="http://numrel.aei.mpg.de/People/personal_webpages/baiotti.html">Tokyo University</a><br />
<!-- <a href="http://www2.polito.it/ricerca/relgrav/">Torino</a><br /> -->
<a href="http://research.physics.uiuc.edu/CTA/IRG/">UIUC</a><br />
<a href="http://www.nuclecu.unam.mx/~gravit/Gravit/">UNAM</a><br />
<!-- <a href="http://cgwa.phys.utb.edu/">UTB</a><br /> -->
<a href="http://wugrav.wustl.edu/">WashU</a><br />
<a href="http://www.yukawa.kyoto-u.ac.jp/english/contents/labs/astro.html">YITP</a></p>
<p><b>Feedback</b><br />
<a href="mailto:schnetter@carpetcode.org">Send email</a></p>
</td>
<td valign="top">
<p>Carpet is an adaptive mesh refinement and multi-patch driver for
the <a href="http://www.cactuscode.org/">Cactus Framework</a>.
Cactus is a software framework for solving time-dependent partial
differential equations on block-structured grids, and Carpet acts
as <i>driver layer</i> providing adaptive mesh refinement,
multi-patch capability, as well as parallelisation and efficient
I/O.</p>
<p>Carpet was created in 2001
by <a href="http://www.cct.lsu.edu/~eschnett/">Erik Schnetter</a>
at the <a href="http://www.tat.physik.uni-tuebingen.de/">TAT</a>
(Theoretische Astrophysik Tübingen) and subsequently brought into
production use by Erik Schnetter, Scott Hawley, and Ian Hawke at
the <a href="http://www.aei.mpg.de/">AEI</a> (Max-Planck-Institut
für Gravitationsphysik, Albert-Einstein-Institut). Carpet is
currently maintained at
the <a href="http://www.cct.lsu.edu/">CCT</a> (Center for
Computation & Technology)
at <a href="http://www.lsu.edu/">LSU</a>. These pages describe
Carpet and its current development.</p>
<hr />
<h2>News</h2>
<table><tr><td valign="top">
<p><b>March 30, 2009:</b> We have ported Carpet to
the <a href="http://www-03.ibm.com/systems/deepcomputing/bluegene/">BlueGene/P</a>
architecture, using
the <a href="http://www.alcf.anl.gov/resources/storage.php">Surveyor</a>
system at the <a href="http://www.alcf.anl.gov/">ALCF</a>. The
graph to the right shows preliminary performance and scaling
results, comparing different compilers and options
(<a href="http://gcc.gnu.org/">gcc</a>, <a href="http://www.ibm.com/software/awdtools/xlcpp/">IBM's
XL compilers</a> without OpenMP, and IBM's XL compilers
with <a href="http://www.openmp.org/">OpenMP</a>, which required
reducing the optimisation level). For these benchmarks, the
problem size was reduced to about one eighth of the standard
size, using 13<sup>3</sup> grid points per core. The results
show that Carpet scales fine up to the size of the total machine
(4k cores), but further work on compiler options is
required.</p>
</td><td valign="top">
<p><a href="scaling-surveyor/results-surveyor.pdf"><img
src="scaling-surveyor/results-surveyor.png"
width="180" alt="AMR benchmark results" /></a></p>
</td></tr></table>
<table><tr><td valign="top">
<p><b>March 20, 2009:</b> Carpet can now perform <i>performance
experiments</i> by artificially increasing the size or the
number of MPI messages exchanged between processes. This can
help determine whether either the communication bandwidth or the
communication latency are a bottleneck of a particular
simulation. The figure to the right shows results for the
standard <a href="http://www.cct.lsu.edu/~eschnett/McLachlan/">McLachlan</a>
AMR benchmark run on
the <a href="http://en.wikipedia.org/wiki/Cray_XT4">Cray XT4</a>
<a href="http://www.nics.tennessee.edu/computing-resources/kraken">Kraken</a>, using 25<sup>3</sup> grid points per core. These
results indicate that the additional latency from increasing the
number of messages has no significant effect, and hence the
benchmark is bandwidth limited for this problem size.</p>
</td><td valign="top">
<p><a href="scaling-whatif/results-whatif.pdf"><img
src="scaling-whatif/results-whatif.png"
width="180" alt="AMR benchmark results" /></a></p>
</td></tr></table>
<table><tr><td valign="top">
<p><b>March 16, 2009:</b> Erik Schnetter and Steve Brandt
published a white
paper <a href="http://www.cct.lsu.edu/CCT-TR/CCT-TR-2009-4"><i>Relativistic
Astrophysics on the SiCortex Architecture</i></a>. This paper
expands on a
<a href="http://www.sicortex.com/news_events/campaigns/lsu_webinar">webinar</a>
by Erik and Steve that was hosted
by <a href="http://www.sicortex.com/">SiCortex</a>.</p>
<p>The graph at the right shows Carpet's parallel scalability
using
the <a href="http://www.cct.lsu.edu/~eschnett/McLachlan/">McLachlan</a>
code with nine levels of AMR for a set of current HPC systems.
The results have been rescaled to the architectures' theoretical
single-core peak performance. This makes it possible to compare
Carpet's scalability on different architectures. (It is not
possible to compare the systems' absolute performance in this
figure.)</p>
</td><td valign="top">
<p><a href="sicortex/results-scaled.pdf"><img
src="sicortex/results-scaled.png"
width="180" alt="AMR benchmark results" /></a></p>
</td></tr></table>
<table><tr><td valign="top">
<p><b>November 9, 2008:</b> In the context of
the <a href="http://www.cct.lsu.edu/xirel/">XiRel project</a>,
we re-designed Carpet's communication layer to avoid many
operations that had a cost of O(<var>N</var>), growing linearly
with the number of MPI processes. Such costs are generally not
acceptable when running on several thousand cores, and have to
be reduced e.g. to O(log <var>N</var>). Carpet now stores the
communication schedule (mostly) in a distributed manner,
increasing performance and reducing its memory requirement.
These improvements are currently being tested; preliminary
scaling results are shown in the figure to the right.</p>
</td><td valign="top">
<p><a href="scaling-improved/results-best.pdf"><img
src="scaling-improved/results-best.png"
width="180" alt="AMR benchmark results" /></a></p>
</td></tr></table>
<p><b>June 25, 2008:</b> We are happy to announce
the <a href="http://www.cct.lsu.edu/~eschnett/SimFactory"><i>Simulation
Factory</i></a>, a tool to help access remote HPC systems,
manage source trees, and submit and control simulations. The
Simulation Factory contains a set of abstractions of the tasks
which are necessary to set up and successfully finish numerical
simulations using the Cactus framework. These abstractions hide
tedious low-level management tasks, they capture "best
practices" of experienced users, and they create a log trail
ensuring repeatable and well-documented scientific results.
Using these abstractions, many types of potentially disastrous
user errors are avoided, and different supercomputers can be
used in a uniform manner.</p>
<table><tr><td valign="top">
<p><b>March 29, 2008:</b> We have benchmarked McLachlan, a new
BSSN-type vacuum Einstein code, using Carpet for unigrid and AMR
calculations. We compare several current large machines:
<a href="http://www.nersc.gov/nusers/systems/franklin/">Franklin</a>
(NERSC), <a href="http://www.loni.org/systems/system.php?system=QueenBee">Queen
Bee</a> (LONI),
and <a href="http://www.tacc.utexas.edu/services/userguides/ranger/">Ranger</a>
(TACC).
<!-- These machines have different architectures and
interconnects.--></p>
</td><td valign="top">
<p><a
href="scaling-amr/results-carpet-1lev.pdf"><img
src="scaling-amr/results-carpet-1lev.png" width="180"
alt="Unigrid benchmark results" /></a></p>
</td><td valign="top">
<p><a
href="scaling-amr/results-carpet-9lev.pdf"><img
src="scaling-amr/results-carpet-9lev.png" width="180"
alt="AMR benchmark results" /></a></p>
</td></tr></table>
<p><a href="olds.html"><b>Old News...</b></a></p>
<hr />
<h2>Documentation</h2>
<p>We have accumulated a few pieces of documentation:</p>
<ul>
<li>An <a href="doc/documentation.pdf">introduction</a>
(PDF, 210 kB) to Carpet, as well as a guide to the
first steps for using it. Everybody should have read this.
(This is the same as the Arrangement Guide from the Carpet
sources.)</li>
<li><a
href="http://www.gravity.psu.edu/numrel/people/sperhake_ulrich.html">Ulrich
Sperhake</a> wrote a tutorial outlining the <a
href="doc/first-steps.pdf">first steps</a> (PDF, 130 kB)
that one has to take to install Carpet and run an example
application.</li>
<li>An explanation of the <a href="doc/internals.pdf">internal
workings</a> (PDF, 120 kB) of Carpet. You should read
this if you want to modify Carpet.</li>
<li>An explanation of
<a href="doc/scheduling.pdf">how scheduling works</a>
(PDF, 120 kB) in (PUGH and) Carpet. This may be
useful for setting up mixtures of local and global operations.</li>
<li>The individual Thorn Guides of Carpet. They are available
with the source code. They give details about the thorns' APIs
and user interfaces.</li>
<li>Thanks to <a
href="http://www.stack.nl/~dimitri/doxygen/">Doxygen</a>, we now
have an <a href="doxygen/html/index.html">overview</a> over all
the routines and data structures in Carpet. Most individual
Doxygen tags are still missing, but the extracted documentation
is already very useful. (The online documentation might not
always be up to date; in case of doubt, extract the
documentation yourself.)</li>
</ul>
<hr />
<h2>Interacting with the developers</h2>
<p>Most discussions about Carpet, i.e. user questions, feature
requests, and bug reports, are held on the Carpet developers'
mailing list <a
href="mailto:developers@lists.carpetcode.org">developers@lists.carpetcode.org</a>.
You can subscribe and unsubscribe from our <a
href="http://lists.carpetcode.org/">list management web page</a>.
You will also find the mailing list archive there. We thank <a
href="http://www.tat.physik.uni-tuebingen.de/~kobras/">Daniel
Kobras</a> for managing the mailing list server.</p>
<p>We have started to use <a
href="http://www.bugzilla.org/">Bugzilla</a> to keep track of
requested features or reported bugs in Carpet. You can submit or
comment on issues from our <a
href="http://bugs.carpetcode.org/">Bugzilla pages</a> once you
have created an account there. The old <a
href="feature-requests.html">list of missing features</a> have not
yet been moved over to Bugzilla.</p>
<hr />
<h2>Pretty pictures</h2>
<p>Here are some pretty pictures of simulations that were
performed with Carpet:</p>
<table>
<tr>
<td valign="top" width="10%">
<p><a href="pictures/meudon-lapse-height.png"><img
src="pictures/thumbnail-meudon-lapse-height.png" height="80"
width="80" alt="lapse height field" /></a></p>
</td>
<td valign="top" width="38%">
<p>Cut through a binary black hole system. Height field of the
lapse function (approximately the time dilatation) in a binary
black hole system calculated from Meudon initial data. The
system is cut between the two black holes, so that only one
black hole is visible. The white boxes indicate the hierarchy
of refinement regions.</p>
</td>
<td valign="top" width="4">
</td>
<td valign="top" width="10">
<p><a href="pictures/quadrupole.jpeg"><img
src="pictures/thumbnail-quadrupole.jpeg" height="80" width="80"
alt="quadrupole wave" /></a></p>
</td>
<td valign="top" width="38%">
<p>A quadrupole wave. Two rotating scalar charges create a
quadrupolar wave, mimicking the gravitational wave trail of a
binary black hole system. The small bumps and riddles are
artifacts caused by the discontinuous charge distribution. To
be improved.</p>
</td>
</tr>
<tr>
<td valign="top">
<p><a href="pictures/meudon-lapse-iso.png"><img
src="pictures/thumbnail-meudon-lapse-iso.png" height="80"
width="80" alt="lapse isosurfaces" /></a></p>
</td>
<td valign="top">
<p>Lapse isosurfaces in a binary black hole system. The same
system as above, but the lapse function is rendered as
isosurfaces.</p>
</td>
<td valign="top">
</td>
<td valign="top">
<p><a href="pictures/collapse-vel-x.png"><img
src="pictures/thumbnail-collapse-vel-x.png" height="80"
width="80" alt="velocity component" /></a></p>
</td>
<td valign="top">
<p>A velocity component in a stellar core collapse. The x
component of the fluid velocity in a stellar core collapse.
This simulation was performed by Christian Ott.</p>
</td>
</tr>
<tr>
<td valign="top">
<p><a href="pictures/multipatch-3phi-error.jpeg"><img
src="pictures/thumbnail-multipatch-3phi-error.jpeg" height="80"
width="80" alt="error function" /></a></p>
</td>
<td valign="top">
<p>The error in a multipatch numerical simulation of scalar wave
propagation in a hollow spherical shell. The coarse- and fine-grid
surface show the numerical errors (computed solution - exact solution)
computed at two different resolutions, with the low resolution error
divided by 16. The fact that the two surfaces overlap nicely shows
that the errors scale as the 4th power of the grid resolution.
This simulation was performed by Jonathan Thornburg.</p>
</td>
<td valign="top">
</td>
<td valign="top">
<p><a href="pictures/matter-density.jpeg"><img
src="pictures/thumbnail-matter-density.jpeg" height="80"
width="80" alt="matter density" /></a></p>
</td>
<td valign="top">
<p>The fate of a proto-neutron-star bar-mode deformation.
Matter density at z=0 during the transition from an m=2 deformed star
to an m=1 deformed one. The light on the right is used to emphasizes
the spiral arms which are responsible for a small mass loss.
This simulation was performed by Gian Mario Manca.</p>
</td>
</tr>
</table>
<p>Moving pictures: We can show
a <a href="movies/waveamr.gif">movie</a> (animated gif,
3.3 MB) of a scalar wave equation with adaptive mesh
refinement. The refinement criterion is a very simplistic local
truncation error estimate. We also have
a <a href="movies/bh2.gif">movie</a> (animated gif, 730 kB)
of a moving refinement region tracking a black hole.</p>
<hr />
<h2>Making sense of results</h2>
<p>Three-dimensional time-dependent simulation results are
difficult enough to interpret when the grid is uniform. With mesh
refinement, the sheer amount of available data makes it necessary
to use professional tools to examine the data. This is not only
the case for "big physics runs", where one (should) know in
advance what to expect, but especially during development, where
things do not always go as planned.
Christian Reisswig was kind enough to write
a <a href="http://www.cactuscode.org/Visualization/VisIt/">database
plugin</a> for the visualisation
tool <a href="https://www.llnl.gov/visit/">VisIt</a>. There is
also
an <a href="http://www.cactuscode.org/Visualization/ImportCarpetHDF5">import
module</a> for the visualisation
tool <a href="http://www.research.ibm.com/dx/">OpenDX</a>
available, implemented by
<a href="http://www.aei.mpg.de/~tradke/">Thomas Radke</a>.</p>
<hr />
<h2>Related projects</h2>
<ul>
<li>Physics: <a href="http://www.cactuscode.org/">Cactus</a>, <a
href="http://numrel.aei.mpg.de/">numrel@aei</a>, <a
href="http://www.whiskycode.org/">Whisky</a></li>
<li>I/O: <a href="http://www.hdfgroup.org/HDF5/">HDF5</a>, <a
href="http://www.fiberbundle.net/">F5</a></li>
<li>Visualisation: <a href="http://www.amiravis.com/">Amira</a>,
<a href="https://www.llnl.gov/visit/">VisIt</a>,
(<a href="http://www.opendx.org/">OpenDX</a>)</li>
</ul>
<hr />
<p>
<a href="http://www.xemacs.org/About/created.html"><img
src="cbxSmall.jpg" alt="Created with XEmacs!" height="36"
width="100" /></a>
<a href="http://www.anybrowser.org/campaign/"><img
src="logoab8.png" alt="Best Viewed With Any Browser" height="31"
width="88" /></a>
<a href="http://validator.w3.org/check?uri=referer"><img
src="valid-xhtml10-blue.png" alt="Valid XHTML 1.0!" height="31"
width="88" /></a>
<!--
<script type="text/javascript"
src="http://boastology.com/tools/pr/?style=2&p=1"></script>
-->
</p>
<address><a href="mailto:schnetter@cct.lsu.edu">Erik Schnetter</a></address>
<p>
<!-- Created: Tue Aug 12 12:12:08 CEST 2003 -->
<!-- hhmts start -->
Last modified: Mon Mar 30 2009
<!-- hhmts end -->
</p>
</td>
</tr>
</table>
</body>
</html>
|