-
-
Notifications
You must be signed in to change notification settings - Fork 5
/
StiffMa.prj
254 lines (253 loc) · 13.3 KB
/
StiffMa.prj
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
<deployment-project plugin="plugin.toolbox" plugin-version="1.0">
<configuration build-checksum="233970263" file="C:\Users\Francisco\Dropbox\StiffMa\Code\StiffMa.prj" location="C:\Users\Francisco\Dropbox\StiffMa\Code" name="StiffMa" target="target.toolbox" target-name="Package Toolbox">
<param.appname>StiffMa</param.appname>
<param.authnamewatermark>Francisco Javier Ramírez-Gil</param.authnamewatermark>
<param.email>fjramireg@gmail.com</param.email>
<param.company>Institución Universitaria Pascual Bravo</param.company>
<param.summary>StiffMa: Fast finite element STIFFness MAtrix generation in MATLAB by using GPU computing.</param.summary>
<param.description>StiffMa: Fast finite element STIFFness MAtrix construction in MATLAB
The finite element method (FEM) is a well established numerical technique for solving partial differential equations (PDEs) in a wide range of complex science and engineering applications. This method has two costly operation that are the construction of global matrices and vectors to form the system of linear or nonlinear equations (assemblage), and their solution (solver). Many efforts have been directed to accelerate the solver. However, the assembly stage has been less investigated although it may represent a serious bottleneck in iterative processes such as non-linear and time-dependent phenomena, and in optimization procedures involving FEM with unstructured meshes. Thus, a fast technique for the global FEM matrices construction is proposed herein by using parallel computing on graphics processing units (GPUs). This work focuses on matrices that arise by solving elliptic PDEs, what is commonly known as stiffness matrix. For performance tests, a scalar problem typically represented by the thermal conduction phenomenon and a vector problem represented by the structural elasticity are considered in a three-dimensional (3D) domain. Unstructured meshes with 8-node hexahedral elements are used to discretize the domain. The MATLAB Parallel Computing Toolbox (PCT) is used to program the CUDA code. The stiffness matrix are built with three GPU kernels that are the indices computation, the numerical integration and the global assembly. Symmetry and adequate data precision are used to save memory and runtime. This proposed methodology allows generating global stiffness matrices from meshes with more than 16.3 millions elements in less than 3 seconds for the scalar problem and up to 3.1 millions for the vector one in 6 seconds using an Nvidia Tesla V100 GPU with 16 GB of memory.
Large speedups are obtained compared with a non-optimized CPU code.</param.description>
<param.screenshot>${PROJECT_ROOT}\StiffMa_logo.png</param.screenshot>
<param.version>1.6</param.version>
<param.output>${PROJECT_ROOT}\StiffMa.mltbx</param.output>
<param.products.name>
<item>MATLAB</item>
<item>Parallel Computing Toolbox</item>
</param.products.name>
<param.products.id>
<item>1</item>
<item>80</item>
</param.products.id>
<param.products.version>
<item>9.8</item>
<item>7.2</item>
</param.products.version>
<param.platforms />
<param.guid>0f1270f7-9280-4376-a635-aefdbce7ff1b</param.guid>
<param.exclude.filters>% List files contained in your toolbox folder that you would like to exclude
% from packaging. Excludes should be listed relative to the toolbox folder.
% Some examples of how to specify excludes are provided below:
%
% A single file in the toolbox folder:
% .svn
.gitignore
desktop.ini
%
% A single file in a subfolder of the toolbox folder:
% example/.svn
%
% All files in a subfolder of the toolbox folder:
% example/*
.git/*
.github/*
%
% All files of a certain name in all subfolders of the toolbox folder:
% **/.svn
%
% All files matching a pattern in all subfolders of the toolbox folder:
% **/*.bak
%</param.exclude.filters>
<param.exclude.pcodedmfiles>true</param.exclude.pcodedmfiles>
<param.examples><?xml version="1.0" encoding="utf-8"?>
<examples>
<exampleCategory name="examples">
<example name="runESTiffScalarCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runESTiffScalarCPUvsGPU.html</file>
<file type="main">/doc/examples/runESTiffScalarCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runEStiffScalarOnCPU" type="html">
<file type="source">/doc/examples/html/runEStiffScalarOnCPU.html</file>
<file type="main">/doc/examples/runEStiffScalarOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runEStiffScalarOnGPU" type="html">
<file type="source">/doc/examples/html/runEStiffScalarOnGPU.html</file>
<file type="main">/doc/examples/runEStiffScalarOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runEStiffVectorCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runEStiffVectorCPUvsGPU.html</file>
<file type="main">/doc/examples/runEStiffVectorCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runEStiffVectorOnCPU" type="html">
<file type="source">/doc/examples/html/runEStiffVectorOnCPU.html</file>
<file type="main">/doc/examples/runEStiffVectorOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runEStiffVectorOnGPU" type="html">
<file type="source">/doc/examples/html/runEStiffVectorOnGPU.html</file>
<file type="main">/doc/examples/runEStiffVectorOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexScalarCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runIndexScalarCPUvsGPU.html</file>
<file type="main">/doc/examples/runIndexScalarCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexScalarOnCPU" type="html">
<file type="source">/doc/examples/html/runIndexScalarOnCPU.html</file>
<file type="main">/doc/examples/runIndexScalarOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexScalarOnGPU" type="html">
<file type="source">/doc/examples/html/runIndexScalarOnGPU.html</file>
<file type="main">/doc/examples/runIndexScalarOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexVectorCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runIndexVectorCPUvsGPU.html</file>
<file type="main">/doc/examples/runIndexVectorCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexVectorOnCPU" type="html">
<file type="source">/doc/examples/html/runIndexVectorOnCPU.html</file>
<file type="main">/doc/examples/runIndexVectorOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runIndexVectorOnGPU" type="html">
<file type="source">/doc/examples/html/runIndexVectorOnGPU.html</file>
<file type="main">/doc/examples/runIndexVectorOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runScalarCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runScalarCPUvsGPU.html</file>
<file type="main">/doc/examples/runScalarCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runScalarOnCPU" type="html">
<file type="source">/doc/examples/html/runScalarOnCPU.html</file>
<file type="main">/doc/examples/runScalarOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runScalarOnGPU" type="html">
<file type="source">/doc/examples/html/runScalarOnGPU.html</file>
<file type="main">/doc/examples/runScalarOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runVectorCPUvsGPU" type="html">
<file type="source">/doc/examples/html/runVectorCPUvsGPU.html</file>
<file type="main">/doc/examples/runVectorCPUvsGPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runVectorOnCPU" type="html">
<file type="source">/doc/examples/html/runVectorOnCPU.html</file>
<file type="main">/doc/examples/runVectorOnCPU.mlx</file>
<file type="thumbnail"/>
</example>
<example name="runVectorOnGPU" type="html">
<file type="source">/doc/examples/html/runVectorOnGPU.html</file>
<file type="main">/doc/examples/runVectorOnGPU.mlx</file>
<file type="thumbnail"/>
</example>
</exampleCategory>
<exampleCategory name="doc">
<example name="GettingStarted" type="html">
<file type="source">/doc/html/GettingStarted.html</file>
<file type="main">/doc/GettingStarted.mlx</file>
<file type="thumbnail"/>
</example>
</exampleCategory>
</examples></param.examples>
<param.demosxml />
<param.apps />
<param.registered.apps />
<param.docs>${PROJECT_ROOT}\tbx\doc\info.xml</param.docs>
<param.getting.started.guide>${PROJECT_ROOT}\tbx\doc\GettingStarted.mlx</param.getting.started.guide>
<param.matlabpath.excludes>
<item>/</item>
</param.matlabpath.excludes>
<param.javaclasspath.excludes />
<param.exported.on.package>true</param.exported.on.package>
<param.required.addons>
<requiredaddons>
<requiredAddOn earliest="earliest" fromRepository="true" id="e57df569-4a80-11e4-9553-005056977bd0" include="false" latest="latest">Postprocessing in FEM</requiredAddOn>
</requiredaddons>
</param.required.addons>
<param.matlab.project.id />
<param.matlab.project.name />
<param.release.start>R2015b</param.release.start>
<param.release.end>latest</param.release.end>
<param.release.current.only>false</param.release.current.only>
<param.compatiblity.windows>true</param.compatiblity.windows>
<param.compatiblity.macos>true</param.compatiblity.macos>
<param.compatiblity.linux>true</param.compatiblity.linux>
<param.compatiblity.matlabonline>true</param.compatiblity.matlabonline>
<param.installation.map />
<param.additional.sw.names />
<param.additional.sw.licenses />
<param.additional.sw.win.url />
<param.additional.sw.mac.url />
<param.additional.sw.linux.url />
<unset>
<param.output />
<param.platforms />
<param.exclude.pcodedmfiles />
<param.demosxml />
<param.apps />
<param.registered.apps />
<param.javaclasspath.excludes />
<param.matlab.project.id />
<param.matlab.project.name />
<param.release.current.only />
<param.compatiblity.windows />
<param.compatiblity.macos />
<param.compatiblity.linux />
<param.compatiblity.matlabonline />
<param.additional.sw.names />
<param.additional.sw.licenses />
<param.additional.sw.win.url />
<param.additional.sw.mac.url />
<param.additional.sw.linux.url />
</unset>
<fileset.rootdir>
<file>${PROJECT_ROOT}\tbx</file>
</fileset.rootdir>
<fileset.rootfiles>
<file>${PROJECT_ROOT}\tbx\demos.xml</file>
<file>${PROJECT_ROOT}\tbx\doc</file>
<file>${PROJECT_ROOT}\tbx\StiffMa</file>
</fileset.rootfiles>
<fileset.depfun.included />
<fileset.depfun.excluded />
<fileset.package />
<build-deliverables>
<file location="${PROJECT_ROOT}" name="StiffMa.mltbx" optional="false">C:\Users\Francisco\Dropbox\StiffMa\Code\StiffMa.mltbx</file>
</build-deliverables>
<workflow />
<matlab>
<root>C:\Program Files\MATLAB\R2020a</root>
<toolboxes>
<toolbox name="matlabcoder" />
<toolbox name="gpucoder" />
</toolboxes>
<toolbox>
<matlabcoder>
<enabled>true</enabled>
</matlabcoder>
</toolbox>
<toolbox>
<gpucoder>
<enabled>true</enabled>
</gpucoder>
</toolbox>
</matlab>
<platform>
<unix>false</unix>
<mac>false</mac>
<windows>true</windows>
<win2k>false</win2k>
<winxp>false</winxp>
<vista>false</vista>
<linux>false</linux>
<solaris>false</solaris>
<osver>10.0</osver>
<os32>false</os32>
<os64>true</os64>
<arch>win64</arch>
<matlab>true</matlab>
</platform>
</configuration>
</deployment-project>