<!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>模型参数检查点(Checkpointing) &mdash; PaddlePaddle  文档</title>
  

  
  

  

  
  
    

  

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

  
  
        <link rel="index" title="索引"
              href="../../genindex.html"/>
        <link rel="search" title="搜索" href="../../search.html"/>
    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/> 

  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
  <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">

  
  <header class="site-header">
    <div class="site-logo">
      <a href="/"><img src="../../_static/images/PP_w.png"></a>
    </div>
    <div class="site-nav-links">
      <div class="site-menu">
        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Folk me on Github</a>
        <div class="language-switcher dropdown">
          <a type="button" data-toggle="dropdown">
            <span>English</span>
            <i class="fa fa-angle-up"></i>
            <i class="fa fa-angle-down"></i>
          </a>
          <ul class="dropdown-menu">
            <li><a href="/doc_cn">中文</a></li>
            <li><a href="/doc">English</a></li>
          </ul>
        </div>
        <ul class="site-page-links">
          <li><a href="/">Home</a></li>
        </ul>
      </div>
      <div class="doc-module">
        
        <ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
</ul>

        
<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>
  </header>
  
  <div class="main-content-wrap">

    
    <nav class="doc-menu-vertical" role="navigation">
        
          
          <ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/ubuntu_install_cn.html">Ubuntu部署PaddlePaddle</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
</ul>

        
    </nav>
    
    <section class="doc-content-wrap">

      

 







<div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
      
    <li>模型参数检查点(Checkpointing)</li>
  </ul>
</div>
      
      <div class="wy-nav-content" id="doc-content">
        <div class="rst-content">
          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
           <div itemprop="articleBody">
            
  <div class="section" id="checkpointing">
<span id="checkpointing"></span><h1>模型参数检查点(Checkpointing)<a class="headerlink" href="#checkpointing" title="永久链接至标题">¶</a></h1>
<p>模型数据检查点的实现,可以有效的避免parameter server的单点或多点同时故障。模型参数检查点通过定期向磁盘上保存一份存储在parameter server内存中的模型数据的完整镜像,来保证训练过程可以从中间状态重新启动。在一个不可中断并缺少备份的训练任务中,可以通过阶段性的保存每个parameter server的数据快照(snapshot)到 <strong><em>分布式存储服务</em></strong> 达到容灾的目的,比如每隔10分钟最新的快照,并删除更早的快照。在出现单点故障时,只需要恢复这台节点,或者将这台节点迁移到另一个节点并启动即可恢复训练任务。</p>
<p><img src="src/checkpointing.png" width="500"/></p>
<div class="section" id="">
<span id="id1"></span><h2>快照保存的设计如下:<a class="headerlink" href="#" title="永久链接至标题">¶</a></h2>
<p>说明:</p>
<ul class="simple">
<li>parameter server在集群中启动后,自动挂载分布式存储目录,并把快照保存到这个目录下。</li>
<li><strong><em>注:每个parameter server的检查点各自独立保存,暂时不考虑多个parameter server同步的保存一个特定时间点的全局检查点,因为这样做也没法保证消除随机性。</em></strong></li>
</ul>
<p>检查点保存程序流程:</p>
<ol class="simple">
<li>如果满足条件&#8221;每隔10分钟&#8221;时,parameter server会获取parameters内存的<code class="docutils literal"><span class="pre">read_lock</span></code>,启动一个新的线程开始保存检查点。如果已经正在执行保存检查点的线程,则忽略。由于对parameters的更新需要获取parameters内存的<code class="docutils literal"><span class="pre">write_lock</span></code>,所以在写入快照的过程中,parameter server会暂停参数更新并等待。</li>
<li>parameter server生成一个UUID,向指定的目录中一个新的文件(文件名为此UUID)写入快照数据。在快照写入完成后,计算这个文件的MD5 sum。然后在etcd的<code class="docutils literal"><span class="pre">/checkpoints/[pserver_id]</span></code>中写入json内容:<code class="docutils literal"><span class="pre">{&quot;uuid&quot;:</span> <span class="pre">[UUID],</span> <span class="pre">&quot;md5&quot;,</span> <span class="pre">&quot;MD5</span> <span class="pre">sum&quot;,</span> <span class="pre">&quot;timestamp&quot;:</span> <span class="pre">xxxx}</span></code>。</li>
<li>删除磁盘目录中不是当前uuid的快照文件。</li>
<li>释放对paramters内存的锁定,停止保存检查点的线程。</li>
</ol>
<p>这里需要用户额外注意,在您的实际环境中,训练任务的运行可能会占满trainer和parameter server之间的网络带宽,如果parameter server此时还需要通过网络访问分布式存储以保存快照,可能会造成网络拥塞,而出现阶段性的运行停滞。</p>
</div>
<div class="section" id="">
<span id="id2"></span><h2>从快照恢复<a class="headerlink" href="#" title="永久链接至标题">¶</a></h2>
<p>在parameter server第一次启动或任意时间parameter server故障后被Kubernetes重新启动,则需要回滚到上一个检查点:</p>
<ol class="simple">
<li>从etcd中读取节点:<code class="docutils literal"><span class="pre">/checkpoints/[pserver_id]</span></code>获取最新的检查点的文件uuid</li>
<li>从磁盘文件中加载uuid文件名的检查点快照文件,并加载其中的参数</li>
<li>如果上面两步出现错误,则使用启动参数定义的初始化方法初始化参数</li>
<li>开始提供服务</li>
</ol>
</div>
</div>
<div class="section" id="todo-list">
<span id="todo-list"></span><h1>TODO List<a class="headerlink" href="#todo-list" title="永久链接至标题">¶</a></h1>
<div class="section" id="todo">
<span id="todo"></span><h2>推测执行/加速执行(TODO)<a class="headerlink" href="#todo" title="永久链接至标题">¶</a></h2>
<p>在异构集群中,如果存在某些trainer执行速度过慢会影响整体集群的速度(如图中Trainer 1),此时master将负责启动一个新的Trainer(Accelerate Trainer 2),使用同样的训练数据block。哪个trainer先完成block的训练,则把另一个慢速的kill掉。</p>
</div>
<div class="section" id="">
<span id="id3"></span><h2>动态扩容/缩容<a class="headerlink" href="#" title="永久链接至标题">¶</a></h2>
<p>目前只考虑动态扩容trainer数量,可以减小系统复杂性。</p>
</div>
</div>
<div class="section" id="">
<span id="id4"></span><h1>术语<a class="headerlink" href="#" title="永久链接至标题">¶</a></h1>
<ul class="simple">
<li>model: 指深度学习训练之后得到的所有参数,使用这个神经网络可以完成对新数据的预测</li>
<li>parameters: 神经网络中的参数,包括权重w和偏置b。一个神经网络的模型由大量的参数组成</li>
<li>shard: 分片,通常指将一个整体拆分成多份的其中的一份。</li>
<li>model shard: 将一个神经网络参数拆分成多份,每个shard分别存储在其中一台parameter server之上</li>
<li>parameter block: 多个parameter block构成一个model shard</li>
<li>单点故障: 任意时刻只可能同时有一台服务器故障。由于集群中同时存在两台机器故障的概率极低((平均故障率*平均故障修复时间)^2)只对特殊在线系统考虑两台以上同时故障的容灾。</li>
</ul>
</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,
            SOURCELINK_SUFFIX: ".txt",
        };
    </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="../../_static/translations.js"></script>
      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
       
  

  
  
    <script type="text/javascript" src="../../_static/js/theme.js"></script>
  
  
  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
  <script src="../../_static/js/paddle_doc_init.js"></script> 

</body>
</html>