parameter_server.html



<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
  <meta charset="utf-8">
  
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  
  <title>Design Doc: Parameter Server &mdash; PaddlePaddle  documentation</title>
  

    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
  

        <link rel="index" title="Index"
              href="../../genindex.html"/>
        <link rel="search" title="Search" href="../../search.html"/>
    <link rel="top" title="PaddlePaddle  documentation" href="../../index.html"/> 
<script>
var _hmt = _hmt || [];
(function() {
  var hm = document.createElement("script");
  hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
  var s = document.getElementsByTagName("script")[0]; 
  s.parentNode.insertBefore(hm, s);
})();
</script>


  <script src="../../_static/js/modernizr.min.js"></script>

</head>

<body class="wy-body-for-nav" role="document">

  <div class="wy-grid-for-nav">

    
    <nav data-toggle="wy-nav-shift" class="wy-nav-side">
      <div class="wy-side-scroll">
        <div class="wy-side-nav-search">
          

            <a href="../../index_en.html" class="icon icon-home"> PaddlePaddle
          

          </a>

          
<div role="search">
  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
    <input type="text" name="q" placeholder="Search docs" />
    <input type="hidden" name="check_keywords" value="yes" />
    <input type="hidden" name="area" value="default" />
  </form>
</div>

          
        </div>

        <div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
          
<nav class="doc-menu-vertical" role="navigation">

<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_en.html">GET STARTED</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/quickstart_en.html">Quick Start</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_en.html">Basic Concept</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../build_and_install/index_en.html">Install and Build</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/pip_install_en.html">Install using pip</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/docker_install_en.html">Run in Docker Containers</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../build_and_install/build_from_source_en.html">Build from Sources</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_en.html">HOW TO</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../howto/cmd_parameter/index_en.html">Set Command-line Parameters</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/use_case_en.html">Use Case</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/arguments_en.html">Argument Outline</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cmd_parameter/detail_introduction_en.html">Detail Description</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/cluster/index_en.html">Distributed Training</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/preparations_en.html">Preparations</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/cmd_argument_en.html">Command-line arguments</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/cluster/multi_cluster/index_en.html">Use different clusters</a><ul>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/fabric_en.html">Fabric</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/openmpi_en.html">OpenMPI</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/k8s_en.html">Kubernetes</a></li>
<li class="toctree-l4"><a class="reference internal" href="../../howto/cluster/multi_cluster/k8s_aws_en.html">Kubernetes on AWS</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/rnn/index_en.html">RNN Models</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/rnn_config_en.html">RNN Configuration</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/recurrent_group_en.html">Recurrent Group Tutorial</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/hierarchical_layer_en.html">Layers supporting hierarchical sequence as input</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/rnn/hrnn_rnn_api_compare_en.html">API comparision between RNN and hierarchical RNN</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_en.html">Tune GPU Performance</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../dev/index_en.html">Development</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../dev/contribute_to_paddle_en.html">Contribute Code</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../dev/write_docs_en.html">Contribute Documentation</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_en.html">FAQ</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_en.html">Install, Build and Unit test</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_en.html">Model Configuration</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_en.html">Parameter Setting</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_en.html">Local Training and Prediction</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_en.html">Cluster Training and Prediction</a></li>
</ul>
</li>
</ul>

</nav>

        </div>
      </div>
    </nav>

    <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">

      
      <nav class="wy-nav-top" role="navigation" aria-label="top navigation">
        <i data-toggle="wy-nav-top" class="fa fa-bars"></i>
        <a href="../../index_en.html">PaddlePaddle</a>
      </nav>


      <div class="wy-nav-content">
        <div class="rst-content">
          

<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
    <li><a href="../../index_en.html">Docs</a> &raquo;</li>
      
    <li>Design Doc: Parameter Server</li>
      <li class="wy-breadcrumbs-aside">
        
          
            <a href="../../_sources/design/dist_refactor/parameter_server.md.txt" rel="nofollow"> View page source</a>
          
        
      </li>
  </ul>
  <hr/>
</div>
          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
           <div itemprop="articleBody">
            
  <div class="section" id="design-doc-parameter-server">
<span id="design-doc-parameter-server"></span><h1>Design Doc: Parameter Server<a class="headerlink" href="#design-doc-parameter-server" title="Permalink to this headline">¶</a></h1>
<div class="section" id="abstract">
<span id="abstract"></span><h2>Abstract<a class="headerlink" href="#abstract" title="Permalink to this headline">¶</a></h2>
<p>We propose an approach to implement the parameter server. In this
approach, there is no fundamental difference between the trainer and
the parameter server: they both run subgraphs, but subgraphs of
different purposes.</p>
</div>
<div class="section" id="background">
<span id="background"></span><h2>Background<a class="headerlink" href="#background" title="Permalink to this headline">¶</a></h2>
<p>The previous implementations of the parameter server do not run a
fluid sub-program. Parameter initialization, optimizer computation, network
communication and checkpointing are implemented twice on both the
trainer as well as the parameter server.</p>
<p>It would be great if we can write code once and use them on both: the
trainer and the parameter server, since this reduces code duplication and
improves extensibility. Given that after the current refactoring, we are
representing everything as a computation graph on the
trainer. Representing everything as a computation graph on the parameter
server becomes a natural extension.</p>
</div>
<div class="section" id="design">
<span id="design"></span><h2>Design<a class="headerlink" href="#design" title="Permalink to this headline">¶</a></h2>
<div class="section" id="distributed-transpiler">
<span id="distributed-transpiler"></span><h3>Distributed Transpiler<a class="headerlink" href="#distributed-transpiler" title="Permalink to this headline">¶</a></h3>
<p>The <em>Distributed Transpiler</em> converts the user-defined fluid program
into sub-programs to be scheduled on different nodes with the following
steps:</p>
<ol class="simple">
<li>OP placement: the OPs will be placed on different nodes according
to a heuristic that minimizes the estimated total computation
time. Currently we will use a simple heuristic that puts parameter
variable on parameter server workers and everything else on trainer
workers.</li>
<li>Add communication OPs to enable the communication between nodes.</li>
</ol>
<p>We will need these OPs: <em>Send</em>, <em>Recv</em>, <em>Enqueue</em>, <em>Dequeue</em>.</p>
<p>Below is an example of converting the user defined graph to the
subgraphs for the trainer and the parameter server:</p>
<p><img src="src/local-graph.png" width="300"/></p>
<p>After converting:</p>
<p><img src="src/dist-graph.png" width="700"/></p>
<ol class="simple">
<li>The parameter variable W and its optimizer program are placed on the parameter server.</li>
<li>Operators are added to the program.<ul>
<li><em>Send</em> sends data to the connected <em>Recv</em> operator.  The
scheduler on the receive node will only schedule <em>Recv</em> operator
to run when the <em>Send</em> operator has ran (the <em>Send</em> OP will mark
the <em>Recv</em> OP runnable automatically).</li>
<li><em>Enqueue</em> enqueues the input variable, it can block until space
become available in the queue.</li>
<li><em>Dequeue</em> outputs configurable numbers of tensors from the
queue. It will block until the queue has the required number of
tensors.</li>
</ul>
</li>
</ol>
</div>
<div class="section" id="benefits">
<span id="benefits"></span><h3>Benefits<a class="headerlink" href="#benefits" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>Model parallelism becomes easier to implement: it is an extension to
the trainer - parameter server approach. We can have several &#8220;Transpilers&#8221;
to achieve different goals.</li>
<li>User-defined optimizer is easier to add - user can now express it as
a sub-program.</li>
<li>No more duplication logic inside the trainer and the parameter
server mentioned in the background section.</li>
</ul>
</div>
<div class="section" id="challenges">
<span id="challenges"></span><h3>Challenges<a class="headerlink" href="#challenges" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>It is important to balance the parameter shards on multiple
parameter servers. If a single parameter is very big (for example: some
word-embedding, fully connected, softmax layer), we need to
automatically partition the single parameter onto different
parameter servers when possible (only element-wise optimizer depends
on the parameter variable).</li>
<li>In the &#8220;Async SGD&#8221; figure, the &#8220;W&#8221; variable on the parameter server
could be read and written concurrently. See
<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/pull/6394">here</a> for more
details about concurrent program in Fluid.</li>
</ul>
</div>
<div class="section" id="discussion">
<span id="discussion"></span><h3>Discussion<a class="headerlink" href="#discussion" title="Permalink to this headline">¶</a></h3>
<ul class="simple">
<li>Can the Enqueue OP be implemented under our current tensor design
(put the input tensor into the queue tensor)?</li>
<li><em>Dequeue</em> OP will have variable numbers of output (depending on the
<code class="docutils literal"><span class="pre">min_count</span></code> attribute), does our current design support it? (similar
question for the <em>Add</em> OP)</li>
</ul>
</div>
<div class="section" id="references">
<span id="references"></span><h3>References:<a class="headerlink" href="#references" title="Permalink to this headline">¶</a></h3>
<p>[1] <a class="reference external" href="https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/45166.pdf">TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems</a></p>
</div>
</div>
</div>


           </div>
          </div>
          <footer>
  

  <hr/>

  <div role="contentinfo">
    <p>
        &copy; Copyright 2016, PaddlePaddle developers.

    </p>
  </div>
  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 

</footer>

        </div>
      </div>

    </section>

  </div>
  

    <script type="text/javascript">
        var DOCUMENTATION_OPTIONS = {
            URL_ROOT:'../../',
            VERSION:'',
            COLLAPSE_INDEX:false,
            FILE_SUFFIX:'.html',
            HAS_SOURCE:  true
        };
    </script>
      <script type="text/javascript" src="../../_static/jquery.js"></script>
      <script type="text/javascript" src="../../_static/underscore.js"></script>
      <script type="text/javascript" src="../../_static/doctools.js"></script>
      <script type="text/javascript" src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script>

  
    <script type="text/javascript" src="../../_static/js/theme.js"></script>
  

  <script type="text/javascript">
      jQuery(function () {
          SphinxRtdTheme.StickyNav.enable();
      });
  </script>
   

</body>
</html>