1<html> 2<head> 3<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> 4<title>Using OpenCL via VexCL</title> 5<link rel="stylesheet" href="../../../../../../../doc/src/boostbook.css" type="text/css"> 6<meta name="generator" content="DocBook XSL Stylesheets V1.79.1"> 7<link rel="home" href="../../index.html" title="Chapter 1. Boost.Numeric.Odeint"> 8<link rel="up" href="../tutorial.html" title="Tutorial"> 9<link rel="prev" href="using_cuda__or_openmp__tbb_______via_thrust.html" title="Using CUDA (or OpenMP, TBB, ...) via Thrust"> 10<link rel="next" href="parallel_computation_with_openmp_and_mpi.html" title="Parallel computation with OpenMP and MPI"> 11</head> 12<body bgcolor="white" text="black" link="#0000FF" vlink="#840084" alink="#0000FF"> 13<table cellpadding="2" width="100%"><tr> 14<td valign="top"><img alt="Boost C++ Libraries" width="277" height="86" src="../../logo.jpg"></td> 15<td align="center"><a href="../../../../../../../index.html">Home</a></td> 16<td align="center"><a href="../../../../../../../libs/libraries.htm">Libraries</a></td> 17<td align="center"><a href="http://www.boost.org/users/people.html">People</a></td> 18<td align="center"><a href="http://www.boost.org/users/faq.html">FAQ</a></td> 19<td align="center"><a href="../../../../../../../more/index.htm">More</a></td> 20</tr></table> 21<hr> 22<div class="spirit-nav"> 23<a accesskey="p" href="using_cuda__or_openmp__tbb_______via_thrust.html"><img src="../../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../tutorial.html"><img src="../../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="parallel_computation_with_openmp_and_mpi.html"><img src="../../../../../../../doc/src/images/next.png" alt="Next"></a> 24</div> 25<div class="section"> 26<div class="titlepage"><div><div><h3 class="title"> 27<a name="boost_numeric_odeint.tutorial.using_opencl_via_vexcl"></a><a class="link" href="using_opencl_via_vexcl.html" title="Using OpenCL via VexCL">Using 28 OpenCL via VexCL</a> 29</h3></div></div></div> 30<p> 31 In the previous section the usage of odeint in combination with <a href="http://code.google.com/p/thrust/" target="_top">Thrust</a> 32 was shown. In this section we show how one can use OpenCL with odeint. The 33 point of odeint is not to implement its own low-level data structures and 34 algorithms, but to use high level libraries doing this task. Here, we will 35 use the <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> framework 36 to use OpenCL. <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 37 is a nice library for general computations and it uses heavily expression 38 templates. With the help of <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 39 it is possible to write very compact and expressive application. 40 </p> 41<div class="note"><table border="0" summary="Note"> 42<tr> 43<td rowspan="2" align="center" valign="top" width="25"><img alt="[Note]" src="../../../../../../../doc/src/images/note.png"></td> 44<th align="left">Note</th> 45</tr> 46<tr><td align="left" valign="top"><p> 47 vexcl needs C++11 features! So you have to compile with C++11 support enabled. 48 </p></td></tr> 49</table></div> 50<p> 51 To use <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> one needs 52 to include one additional header which includes the data-types and algorithms 53 from vexcl and the adaption to odeint. Adaption to odeint means here only 54 to adapt the resizing functionality of <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 55 to odeint. 56 </p> 57<p> 58</p> 59<pre class="programlisting"><span class="preprocessor">#include</span> <span class="special"><</span><span class="identifier">boost</span><span class="special">/</span><span class="identifier">numeric</span><span class="special">/</span><span class="identifier">odeint</span><span class="special">/</span><span class="identifier">external</span><span class="special">/</span><span class="identifier">vexcl</span><span class="special">/</span><span class="identifier">vexcl</span><span class="special">.</span><span class="identifier">hpp</span><span class="special">></span> 60</pre> 61<p> 62 </p> 63<p> 64 To demonstrate the use of <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 65 we integrate an ensemble of Lorenz system. The example is very similar to 66 the parameter study of the Lorenz system in the previous section except that 67 we do not compute the Lyapunov exponents. Again, we vary the parameter R 68 of the Lorenz system an solve a whole ensemble of Lorenz systems in parallel 69 (each with a different parameter R). First, we define the state type and 70 a vector type 71 </p> 72<p> 73</p> 74<pre class="programlisting"><span class="keyword">typedef</span> <span class="identifier">vex</span><span class="special">::</span><span class="identifier">vector</span><span class="special"><</span> <span class="keyword">double</span> <span class="special">></span> <span class="identifier">vector_type</span><span class="special">;</span> 75<span class="keyword">typedef</span> <span class="identifier">vex</span><span class="special">::</span><span class="identifier">multivector</span><span class="special"><</span> <span class="keyword">double</span><span class="special">,</span> <span class="number">3</span> <span class="special">></span> <span class="identifier">state_type</span><span class="special">;</span> 76</pre> 77<p> 78 </p> 79<p> 80 The <code class="computeroutput"><span class="identifier">vector_type</span></code> is used to 81 represent the parameter R. The <code class="computeroutput"><span class="identifier">state_type</span></code> 82 is a multi-vector of three sub vectors and is used to represent. The first 83 component of this multi-vector represent all <code class="computeroutput"><span class="identifier">x</span></code> 84 components of the Lorenz system, while the second all <code class="computeroutput"><span class="identifier">y</span></code> 85 components and the third all <code class="computeroutput"><span class="identifier">z</span></code> 86 components. The components of this vector can be obtained via 87 </p> 88<p> 89</p> 90<pre class="programlisting"><span class="keyword">auto</span> <span class="special">&</span><span class="identifier">x</span> <span class="special">=</span> <span class="identifier">X</span><span class="special">(</span><span class="number">0</span><span class="special">);</span> 91<span class="keyword">auto</span> <span class="special">&</span><span class="identifier">y</span> <span class="special">=</span> <span class="identifier">X</span><span class="special">(</span><span class="number">1</span><span class="special">);</span> 92<span class="keyword">auto</span> <span class="special">&</span><span class="identifier">z</span> <span class="special">=</span> <span class="identifier">X</span><span class="special">(</span><span class="number">2</span><span class="special">);</span> 93</pre> 94<p> 95 </p> 96<p> 97 As already mentioned <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 98 supports expression templates and we will use them to implement the system 99 function for the Lorenz ensemble: 100 </p> 101<p> 102</p> 103<pre class="programlisting"><span class="keyword">const</span> <span class="keyword">double</span> <span class="identifier">sigma</span> <span class="special">=</span> <span class="number">10.0</span><span class="special">;</span> 104<span class="keyword">const</span> <span class="keyword">double</span> <span class="identifier">b</span> <span class="special">=</span> <span class="number">8.0</span> <span class="special">/</span> <span class="number">3.0</span><span class="special">;</span> 105 106<span class="keyword">struct</span> <span class="identifier">sys_func</span> 107<span class="special">{</span> 108 <span class="keyword">const</span> <span class="identifier">vector_type</span> <span class="special">&</span><span class="identifier">R</span><span class="special">;</span> 109 110 <span class="identifier">sys_func</span><span class="special">(</span> <span class="keyword">const</span> <span class="identifier">vector_type</span> <span class="special">&</span><span class="identifier">_R</span> <span class="special">)</span> <span class="special">:</span> <span class="identifier">R</span><span class="special">(</span> <span class="identifier">_R</span> <span class="special">)</span> <span class="special">{</span> <span class="special">}</span> 111 112 <span class="keyword">void</span> <span class="keyword">operator</span><span class="special">()(</span> <span class="keyword">const</span> <span class="identifier">state_type</span> <span class="special">&</span><span class="identifier">x</span> <span class="special">,</span> <span class="identifier">state_type</span> <span class="special">&</span><span class="identifier">dxdt</span> <span class="special">,</span> <span class="keyword">double</span> <span class="identifier">t</span> <span class="special">)</span> <span class="keyword">const</span> 113 <span class="special">{</span> 114 <span class="identifier">dxdt</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">=</span> <span class="special">-</span><span class="identifier">sigma</span> <span class="special">*</span> <span class="special">(</span> <span class="identifier">x</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">-</span> <span class="identifier">x</span><span class="special">(</span><span class="number">1</span><span class="special">)</span> <span class="special">);</span> 115 <span class="identifier">dxdt</span><span class="special">(</span><span class="number">1</span><span class="special">)</span> <span class="special">=</span> <span class="identifier">R</span> <span class="special">*</span> <span class="identifier">x</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">-</span> <span class="identifier">x</span><span class="special">(</span><span class="number">1</span><span class="special">)</span> <span class="special">-</span> <span class="identifier">x</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">*</span> <span class="identifier">x</span><span class="special">(</span><span class="number">2</span><span class="special">);</span> 116 <span class="identifier">dxdt</span><span class="special">(</span><span class="number">2</span><span class="special">)</span> <span class="special">=</span> <span class="special">-</span> <span class="identifier">b</span> <span class="special">*</span> <span class="identifier">x</span><span class="special">(</span><span class="number">2</span><span class="special">)</span> <span class="special">+</span> <span class="identifier">x</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">*</span> <span class="identifier">x</span><span class="special">(</span><span class="number">1</span><span class="special">);</span> 117 <span class="special">}</span> 118<span class="special">};</span> 119</pre> 120<p> 121 </p> 122<p> 123 It's very easy, isn't it? These three little lines do all the computations 124 for you. There is no need to write your own OpenCL kernels. <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 125 does everything for you. Next we have to write the main application. We initialize 126 the vector of parameters (R) and the initial state. Note that <a href="https://github.com/ddemidov/vexcl" target="_top">VexCL</a> 127 requires the <code class="computeroutput"><span class="identifier">vector_space_algebra</span></code>, 128 but that is automatically deduced and configured by odeint internally, so 129 we only have to specify the <code class="computeroutput"><span class="identifier">state_type</span></code> 130 when instantiating the stepper and we are done: 131 </p> 132<p> 133</p> 134<pre class="programlisting"><span class="comment">// setup the opencl context</span> 135<span class="identifier">vex</span><span class="special">::</span><span class="identifier">Context</span> <span class="identifier">ctx</span><span class="special">(</span> <span class="identifier">vex</span><span class="special">::</span><span class="identifier">Filter</span><span class="special">::</span><span class="identifier">Type</span><span class="special">(</span><span class="identifier">CL_DEVICE_TYPE_GPU</span><span class="special">)</span> <span class="special">);</span> 136<span class="identifier">std</span><span class="special">::</span><span class="identifier">cout</span> <span class="special"><<</span> <span class="identifier">ctx</span> <span class="special"><<</span> <span class="identifier">std</span><span class="special">::</span><span class="identifier">endl</span><span class="special">;</span> 137 138<span class="comment">// set up number of system, time step and integration time</span> 139<span class="keyword">const</span> <span class="identifier">size_t</span> <span class="identifier">n</span> <span class="special">=</span> <span class="number">1024</span> <span class="special">*</span> <span class="number">1024</span><span class="special">;</span> 140<span class="keyword">const</span> <span class="keyword">double</span> <span class="identifier">dt</span> <span class="special">=</span> <span class="number">0.01</span><span class="special">;</span> 141<span class="keyword">const</span> <span class="keyword">double</span> <span class="identifier">t_max</span> <span class="special">=</span> <span class="number">1000.0</span><span class="special">;</span> 142 143<span class="comment">// initialize R</span> 144<span class="keyword">double</span> <span class="identifier">Rmin</span> <span class="special">=</span> <span class="number">0.1</span> <span class="special">,</span> <span class="identifier">Rmax</span> <span class="special">=</span> <span class="number">50.0</span> <span class="special">,</span> <span class="identifier">dR</span> <span class="special">=</span> <span class="special">(</span> <span class="identifier">Rmax</span> <span class="special">-</span> <span class="identifier">Rmin</span> <span class="special">)</span> <span class="special">/</span> <span class="keyword">double</span><span class="special">(</span> <span class="identifier">n</span> <span class="special">-</span> <span class="number">1</span> <span class="special">);</span> 145<span class="identifier">std</span><span class="special">::</span><span class="identifier">vector</span><span class="special"><</span><span class="keyword">double</span><span class="special">></span> <span class="identifier">x</span><span class="special">(</span> <span class="identifier">n</span> <span class="special">*</span> <span class="number">3</span> <span class="special">)</span> <span class="special">,</span> <span class="identifier">r</span><span class="special">(</span> <span class="identifier">n</span> <span class="special">);</span> 146<span class="keyword">for</span><span class="special">(</span> <span class="identifier">size_t</span> <span class="identifier">i</span><span class="special">=</span><span class="number">0</span> <span class="special">;</span> <span class="identifier">i</span><span class="special"><</span><span class="identifier">n</span> <span class="special">;</span> <span class="special">++</span><span class="identifier">i</span> <span class="special">)</span> <span class="identifier">r</span><span class="special">[</span><span class="identifier">i</span><span class="special">]</span> <span class="special">=</span> <span class="identifier">Rmin</span> <span class="special">+</span> <span class="identifier">dR</span> <span class="special">*</span> <span class="keyword">double</span><span class="special">(</span> <span class="identifier">i</span> <span class="special">);</span> 147<span class="identifier">vector_type</span> <span class="identifier">R</span><span class="special">(</span> <span class="identifier">ctx</span><span class="special">.</span><span class="identifier">queue</span><span class="special">()</span> <span class="special">,</span> <span class="identifier">r</span> <span class="special">);</span> 148 149<span class="comment">// initialize the state of the lorenz ensemble</span> 150<span class="identifier">state_type</span> <span class="identifier">X</span><span class="special">(</span><span class="identifier">ctx</span><span class="special">.</span><span class="identifier">queue</span><span class="special">(),</span> <span class="identifier">n</span><span class="special">);</span> 151<span class="identifier">X</span><span class="special">(</span><span class="number">0</span><span class="special">)</span> <span class="special">=</span> <span class="number">10.0</span><span class="special">;</span> 152<span class="identifier">X</span><span class="special">(</span><span class="number">1</span><span class="special">)</span> <span class="special">=</span> <span class="number">10.0</span><span class="special">;</span> 153<span class="identifier">X</span><span class="special">(</span><span class="number">2</span><span class="special">)</span> <span class="special">=</span> <span class="number">10.0</span><span class="special">;</span> 154 155<span class="comment">// create a stepper</span> 156<span class="identifier">runge_kutta4</span><span class="special"><</span> <span class="identifier">state_type</span> <span class="special">></span> <span class="identifier">stepper</span><span class="special">;</span> 157 158<span class="comment">// solve the system</span> 159<span class="identifier">integrate_const</span><span class="special">(</span> <span class="identifier">stepper</span> <span class="special">,</span> <span class="identifier">sys_func</span><span class="special">(</span> <span class="identifier">R</span> <span class="special">)</span> <span class="special">,</span> <span class="identifier">X</span> <span class="special">,</span> <span class="number">0.0</span> <span class="special">,</span> <span class="identifier">t_max</span> <span class="special">,</span> <span class="identifier">dt</span> <span class="special">);</span> 160</pre> 161<p> 162 </p> 163</div> 164<table xmlns:rev="http://www.cs.rpi.edu/~gregod/boost/tools/doc/revision" width="100%"><tr> 165<td align="left"></td> 166<td align="right"><div class="copyright-footer">Copyright © 2009-2015 Karsten Ahnert and Mario Mulansky<p> 167 Distributed under the Boost Software License, Version 1.0. (See accompanying 168 file LICENSE_1_0.txt or copy at <a href="http://www.boost.org/LICENSE_1_0.txt" target="_top">http://www.boost.org/LICENSE_1_0.txt</a>) 169 </p> 170</div></td> 171</tr></table> 172<hr> 173<div class="spirit-nav"> 174<a accesskey="p" href="using_cuda__or_openmp__tbb_______via_thrust.html"><img src="../../../../../../../doc/src/images/prev.png" alt="Prev"></a><a accesskey="u" href="../tutorial.html"><img src="../../../../../../../doc/src/images/up.png" alt="Up"></a><a accesskey="h" href="../../index.html"><img src="../../../../../../../doc/src/images/home.png" alt="Home"></a><a accesskey="n" href="parallel_computation_with_openmp_and_mpi.html"><img src="../../../../../../../doc/src/images/next.png" alt="Next"></a> 175</div> 176</body> 177</html> 178
