|
|
<!DOCTYPE html>
<!-- Generated by pkgdown: do not edit by hand --><html><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"><meta name="viewport" content="width=device-width, initial-scale=1.0"><title>Performance benefit by using compiled model definitions in mkin • mkin</title><!-- jquery --><script src="https://code.jquery.com/jquery-3.1.0.min.js" integrity="sha384-nrOSfDHtoPMzJHjVTdCopGqIqeYETSXhZDFyniQ8ZHcVy08QesyHcnOUpMpqnmWq" crossorigin="anonymous"></script><!-- Bootstrap --><link href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-BVYiiSIFeK1dGmJRAkycuHAHRg32OmUcww7on3RYdg4Va+PmSTsz/K68vbdEjh4u" crossorigin="anonymous"><script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script><!-- Font Awesome icons --><link href="https://maxcdn.bootstrapcdn.com/font-awesome/4.6.3/css/font-awesome.min.css" rel="stylesheet" integrity="sha384-T8Gy5hrqNKT+hzMclPo118YTQO6cYprQmhrYwIiQ/3axmI1hQomh7Ud2hPOy8SP1" crossorigin="anonymous"><!-- pkgdown --><link href="../pkgdown.css" rel="stylesheet"><script src="../jquery.sticky-kit.min.js"></script><script src="../pkgdown.js"></script><!-- mathjax --><script src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script><!--[if lt IE 9]>
<script src="https://oss.maxcdn.com/html5shiv/3.7.3/html5shiv.min.js"></script>
<script src="https://oss.maxcdn.com/respond/1.4.2/respond.min.js"></script>
<![endif]--></head><body>
<div class="container template-vignette">
<header><div class="navbar navbar-default navbar-fixed-top" role="navigation">
<div class="container">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar">
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a class="navbar-brand" href="../index.html">mkin</a>
</div>
<div id="navbar" class="navbar-collapse collapse">
<ul class="nav navbar-nav"><li>
<a href="../reference/index.html">Reference</a>
</li>
<li>
<a href="../articles/index.html">Articles</a>
</li>
<li>
<a href="../news/index.html">News</a>
</li>
</ul><ul class="nav navbar-nav navbar-right"><li>
<a href="http://github.com/jranke/mkin">
<span class="fa fa-github fa-lg"></span>
</a>
</li>
</ul></div><!--/.nav-collapse -->
</div><!--/.container -->
</div><!--/.navbar -->
</header><div class="row">
<div class="col-md-9">
<div class="page-header toc-ignore">
<h1>Performance benefit by using compiled model definitions in mkin</h1>
<h4 class="author">Johannes Ranke</h4>
<h4 class="date">2016-11-18</h4>
</div>
<div class="contents">
<div id="model-that-can-also-be-solved-with-eigenvalues" class="section level2">
<h2 class="hasAnchor"><html><body><a href="#model-that-can-also-be-solved-with-eigenvalues" class="anchor"> </a></body></html>Model that can also be solved with Eigenvalues</h2>
<p>This evaluation is taken from the example section of mkinfit. When using an mkin version equal to or greater than 0.9-36 and a C compiler (gcc) is available, you will see a message that the model is being compiled from autogenerated C code when defining a model using mkinmod. The <code><a href="../reference/mkinmod.html">mkinmod()</a></code> function checks for presence of the gcc compiler using</p>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">Sys.which</span>(<span class="st">"gcc"</span>)</code></pre></div>
<pre><code>## gcc
## "/usr/bin/gcc"</code></pre>
<p>First, we build a simple degradation model for a parent compound with one metabolite.</p>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">library</span>(<span class="st">"mkin"</span>)
SFO_SFO <-<span class="st"> </span><span class="kw"><a href="../reference/mkinmod.html">mkinmod</a></span>(
<span class="dt">parent =</span> <span class="kw"><a href="../reference/mkinsub.html">mkinsub</a></span>(<span class="st">"SFO"</span>, <span class="st">"m1"</span>),
<span class="dt">m1 =</span> <span class="kw"><a href="../reference/mkinsub.html">mkinsub</a></span>(<span class="st">"SFO"</span>))</code></pre></div>
<pre><code>## Successfully compiled differential equation model from auto-generated C code.</code></pre>
<p>We can compare the performance of the Eigenvalue based solution against the compiled version and the R implementation of the differential equations using the microbenchmark package.</p>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">library</span>(<span class="st">"microbenchmark"</span>)
<span class="kw">library</span>(<span class="st">"ggplot2"</span>)
mb<span class="fl">.1</span> <-<span class="st"> </span><span class="kw">microbenchmark</span>(
<span class="st">"deSolve, not compiled"</span> =<span class="st"> </span><span class="kw"><a href="../reference/mkinfit.html">mkinfit</a></span>(SFO_SFO, FOCUS_2006_D,
<span class="dt">solution_type =</span> <span class="st">"deSolve"</span>,
<span class="dt">use_compiled =</span> <span class="ot">FALSE</span>, <span class="dt">quiet =</span> <span class="ot">TRUE</span>),
<span class="st">"Eigenvalue based"</span> =<span class="st"> </span><span class="kw"><a href="../reference/mkinfit.html">mkinfit</a></span>(SFO_SFO, FOCUS_2006_D,
<span class="dt">solution_type =</span> <span class="st">"eigen"</span>, <span class="dt">quiet =</span> <span class="ot">TRUE</span>),
<span class="st">"deSolve, compiled"</span> =<span class="st"> </span><span class="kw"><a href="../reference/mkinfit.html">mkinfit</a></span>(SFO_SFO, FOCUS_2006_D,
<span class="dt">solution_type =</span> <span class="st">"deSolve"</span>, <span class="dt">quiet =</span> <span class="ot">TRUE</span>),
<span class="dt">times =</span> <span class="dv">3</span>, <span class="dt">control =</span> <span class="kw">list</span>(<span class="dt">warmup =</span> <span class="dv">0</span>))</code></pre></div>
<pre><code>## Warning in microbenchmark(`deSolve, not compiled` = mkinfit(SFO_SFO,
## FOCUS_2006_D, : Could not measure overhead. Your clock might lack
## precision.</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">smb<span class="fl">.1</span> <-<span class="st"> </span><span class="kw">summary</span>(mb<span class="fl">.1</span>)
<span class="kw">print</span>(mb<span class="fl">.1</span>)</code></pre></div>
<pre><code>## Unit: milliseconds
## expr min lq mean median uq
## deSolve, not compiled 6341.1795 6363.5977 6373.7748 6386.0158 6390.0724
## Eigenvalue based 907.1699 908.8498 921.8335 910.5297 929.1653
## deSolve, compiled 747.5311 752.3718 754.2086 757.2125 757.5474
## max neval cld
## 6394.1290 3 c
## 947.8008 3 b
## 757.8823 3 a</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">autoplot</span>(mb<span class="fl">.1</span>)</code></pre></div>
<p><img src="compiled_models_files/figure-html/benchmark_SFO_SFO-1.png" width="672"></p>
<p>We see that using the compiled model is by a factor of 8.4 faster than using the R version with the default ode solver, and it is even faster than the Eigenvalue based solution implemented in R which does not need iterative solution of the ODEs:</p>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">rownames</span>(smb<span class="fl">.1</span>) <-<span class="st"> </span>smb<span class="fl">.1</span>$expr
smb<span class="fl">.1</span>[<span class="st">"median"</span>]/smb<span class="fl">.1</span>[<span class="st">"deSolve, compiled"</span>, <span class="st">"median"</span>]</code></pre></div>
<pre><code>## median
## deSolve, not compiled 8.433585
## Eigenvalue based 1.202476
## deSolve, compiled 1.000000</code></pre>
</div>
<div id="model-that-can-not-be-solved-with-eigenvalues" class="section level2">
<h2 class="hasAnchor"><html><body><a href="#model-that-can-not-be-solved-with-eigenvalues" class="anchor"> </a></body></html>Model that can not be solved with Eigenvalues</h2>
<p>This evaluation is also taken from the example section of mkinfit.</p>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">FOMC_SFO <-<span class="st"> </span><span class="kw"><a href="../reference/mkinmod.html">mkinmod</a></span>(
<span class="dt">parent =</span> <span class="kw"><a href="../reference/mkinsub.html">mkinsub</a></span>(<span class="st">"FOMC"</span>, <span class="st">"m1"</span>),
<span class="dt">m1 =</span> <span class="kw"><a href="../reference/mkinsub.html">mkinsub</a></span>( <span class="st">"SFO"</span>))</code></pre></div>
<pre><code>## Successfully compiled differential equation model from auto-generated C code.</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">mb<span class="fl">.2</span> <-<span class="st"> </span><span class="kw">microbenchmark</span>(
<span class="st">"deSolve, not compiled"</span> =<span class="st"> </span><span class="kw"><a href="../reference/mkinfit.html">mkinfit</a></span>(FOMC_SFO, FOCUS_2006_D,
<span class="dt">use_compiled =</span> <span class="ot">FALSE</span>, <span class="dt">quiet =</span> <span class="ot">TRUE</span>),
<span class="st">"deSolve, compiled"</span> =<span class="st"> </span><span class="kw"><a href="../reference/mkinfit.html">mkinfit</a></span>(FOMC_SFO, FOCUS_2006_D, <span class="dt">quiet =</span> <span class="ot">TRUE</span>),
<span class="dt">times =</span> <span class="dv">3</span>, <span class="dt">control =</span> <span class="kw">list</span>(<span class="dt">warmup =</span> <span class="dv">0</span>))</code></pre></div>
<pre><code>## Warning in microbenchmark(`deSolve, not compiled` = mkinfit(FOMC_SFO,
## FOCUS_2006_D, : Could not measure overhead. Your clock might lack
## precision.</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">smb<span class="fl">.2</span> <-<span class="st"> </span><span class="kw">summary</span>(mb<span class="fl">.2</span>)
<span class="kw">print</span>(mb<span class="fl">.2</span>)</code></pre></div>
<pre><code>## Unit: seconds
## expr min lq mean median uq
## deSolve, not compiled 13.214250 13.248908 13.384577 13.283565 13.469740
## deSolve, compiled 1.303868 1.307484 1.332879 1.311099 1.347384
## max neval cld
## 13.655915 3 b
## 1.383669 3 a</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">smb<span class="fl">.2</span>[<span class="st">"median"</span>]/smb<span class="fl">.2</span>[<span class="st">"deSolve, compiled"</span>, <span class="st">"median"</span>]</code></pre></div>
<pre><code>## median
## 1 NA
## 2 NA</code></pre>
<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">autoplot</span>(mb<span class="fl">.2</span>)</code></pre></div>
<p><img src="compiled_models_files/figure-html/benchmark_FOMC_SFO-1.png" width="672"></p>
<p>Here we get a performance benefit of a factor of 10.1 using the version of the differential equation model compiled from C code!</p>
<p>This vignette was built with mkin 0.9.44.9000 on</p>
<pre><code>## R version 3.3.2 (2016-10-31)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Debian GNU/Linux 8 (jessie)</code></pre>
<pre><code>## CPU model: Intel(R) Core(TM) i7-4710MQ CPU @ 2.50GHz</code></pre>
</div>
</div>
</div>
<div class="col-md-3 hidden-xs hidden-sm" id="sidebar">
<div id="tocnav">
<h2>Contents</h2>
<ul class="nav nav-pills nav-stacked"><li><a href="#model-that-can-also-be-solved-with-eigenvalues">Model that can also be solved with Eigenvalues</a></li>
<li><a href="#model-that-can-not-be-solved-with-eigenvalues">Model that can not be solved with Eigenvalues</a></li>
</ul></div>
</div>
</div>
<footer><div class="copyright">
<p>Developed by Johannes Ranke.</p>
</div>
<div class="pkgdown">
<p>Site built with <a href="http://hadley.github.io/pkgdown/">pkgdown</a>.</p>
</div>
</footer></div>
</body></html>
|
