master

分支 (1)

管理

管理

master

blog
/
Spark项目落地实战以及日常大数据开发注意事项.html

<!DOCTYPE html>


<html class="theme-next pisces use-motion" lang="zh-CN">
<head><meta name="generator" content="Hexo 3.8.0">
  <meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=2">
<meta name="theme-color" content="#222">


    <link rel="stylesheet" href="//fonts.googleapis.com/css?family=Monda:300,300italic,400,400italic,700,700italic|Roboto Slab:300,300italic,400,400italic,700,700italic|Lobster Two:300,300italic,400,400italic,700,700italic|PT Mono:300,300italic,400,400italic,700,700italic&subset=latin,latin-ext">


<link rel="stylesheet" href="/lib/font-awesome/css/font-awesome.min.css?v=4.7.0">

<link rel="stylesheet" href="/css/main.css?v=7.1.2">


  <link rel="apple-touch-icon" sizes="180x180" href="/images/apple-touch-icon-next.png?v=7.1.2">


  <link rel="icon" type="image/png" sizes="32x32" href="/images/favicon-32x32-next.png?v=7.1.2">


  <link rel="icon" type="image/png" sizes="16x16" href="/favicon.ico?v=7.1.2">


  <link rel="mask-icon" href="/images/logo.svg?v=7.1.2" color="#222">


<script id="hexo.configurations">
  var NexT = window.NexT || {};
  var CONFIG = {
    root: '/',
    scheme: 'Pisces',
    version: '7.1.2',
    sidebar: {"position":"left","display":"hide","offset":12,"onmobile":false,"dimmer":false},
    back2top: true,
    back2top_sidebar: false,
    fancybox: false,
    fastclick: false,
    lazyload: false,
    tabs: true,
    motion: {"enable":true,"async":true,"transition":{"post_block":"fadeIn","post_header":"slideDownIn","post_body":"slideDownIn","coll_header":"slideLeftIn","sidebar":"slideUpIn"}},
    algolia: {
      applicationID: '',
      apiKey: '',
      indexName: '',
      hits: {"per_page":10},
      labels: {"input_placeholder":"Search for Posts","hits_empty":"We didn't find any results for the search: ${query}","hits_stats":"${hits} results found in ${time} ms"}
    }
  };
</script>


  <meta name="description" content="Spark简介">
<meta name="keywords" content="spark">
<meta property="og:type" content="article">
<meta property="og:title" content="Spark项目落地实战以及日常大数据开发注意事项">
<meta property="og:url" content="https://www.dudefu.tk/Spark项目落地实战以及日常大数据开发注意事项.html">
<meta property="og:site_name" content="The Future">
<meta property="og:description" content="Spark简介">
<meta property="og:locale" content="zh-CN">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7g9ip7j30fa0770tg.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7iy2qnj30fa08nju4.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7lcf7aj30fa03sgmk.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7p5ynzj30fa078glw.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7scd41j30fa07640c.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7vircej30fa09a750.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7ygkecj30fa09sta6.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks81s5qnj30fa05574q.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks842uowj30fa07bq5d.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks86w7u9j30fa08cac6.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks89mfdaj30fa08k40z.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8dail0j30fa0csjvb.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8g81p6j30fa0chq75.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8j994gj30fa07rta1.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8lykeaj30fa0enwkb.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8p5bh0j30fa0cc78s.jpg">
<meta property="og:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8sw7yuj30fa0bc456.jpg">
<meta property="og:updated_time" content="2021-01-12T04:08:55.595Z">
<meta name="twitter:card" content="summary">
<meta name="twitter:title" content="Spark项目落地实战以及日常大数据开发注意事项">
<meta name="twitter:description" content="Spark简介">
<meta name="twitter:image" content="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7g9ip7j30fa0770tg.jpg">


  <link rel="canonical" href="https://www.dudefu.tk/Spark项目落地实战以及日常大数据开发注意事项">


<script id="page.configurations">
  CONFIG.page = {
    sidebar: "",
  };
</script>

  <title>Spark项目落地实战以及日常大数据开发注意事项 | The Future</title>


  <noscript>
  <style>
  .use-motion .motion-element,
  .use-motion .brand,
  .use-motion .menu-item,
  .sidebar-inner,
  .use-motion .post-block,
  .use-motion .pagination,
  .use-motion .comments,
  .use-motion .post-header,
  .use-motion .post-body,
  .use-motion .collection-title { opacity: initial; }

  .use-motion .logo,
  .use-motion .site-title,
  .use-motion .site-subtitle {
    opacity: initial;
    top: initial;
  }

  .use-motion .logo-line-before i { left: initial; }
  .use-motion .logo-line-after i { right: initial; }
  </style>
</noscript>

</head>

<body itemscope="" itemtype="http://schema.org/WebPage" lang="zh-CN">


  <div class="container sidebar-position-left page-post-detail">
    <div class="headband"></div>

    <header id="header" class="header" itemscope="" itemtype="http://schema.org/WPHeader">
      <div class="header-inner"><div class="site-brand-wrapper">
  <div class="site-meta">


    <div class="custom-logo-site-title">
      <a href="/" class="brand" rel="start">
        <span class="logo-line-before"><i></i></span>
        <span class="site-title">The Future</span>
        <span class="logo-line-after"><i></i></span>
      </a>
    </div>


        <h1 class="site-subtitle" itemprop="description">Stay hungry,stay foolish.</h1>


  </div>

  <div class="site-nav-toggle">
    <button aria-label="切换导航栏">
      <span class="btn-bar"></span>
      <span class="btn-bar"></span>
      <span class="btn-bar"></span>
    </button>
  </div>
</div>


<nav class="site-nav">

    <ul id="menu" class="menu">


          <li class="menu-item menu-item-home">


    <a href="/" rel="section"><i class="menu-item-icon fa fa-fw fa-home"></i> <br>首页</a>

  </li>


          <li class="menu-item menu-item-archives">


    <a href="/archives/" rel="section"><i class="menu-item-icon fa fa-fw fa-archive"></i> <br>归档<span class="badge">125</span></a>

  </li>


          <li class="menu-item menu-item-categories">


    <a href="/categories" rel="section"><i class="menu-item-icon fa fa-fw fa-th"></i> <br>分类<span class="badge">15</span></a>

  </li>


          <li class="menu-item menu-item-tags">


    <a href="/tags" rel="section"><i class="menu-item-icon fa fa-fw fa-tags"></i> <br>标签<span class="badge">63</span></a>

  </li>


          <li class="menu-item menu-item-something">


    <a href="/something" rel="section"><i class="menu-item-icon fa fa-fw fa-paper-plane"></i> <br>干货</a>

  </li>


          <li class="menu-item menu-item-about">


    <a href="/about/" rel="section"><i class="menu-item-icon fa fa-fw fa-user"></i> <br>关于</a>

  </li>


        <li class="menu-item menu-item-search">

            <a href="javascript:;" class="popup-trigger">


              <i class="menu-item-icon fa fa-search fa-fw"></i> <br>搜索</a>
        </li>

    </ul>


    <div class="site-search">

  <div class="popup search-popup local-search-popup">
  <div class="local-search-header clearfix">
    <span class="search-icon">
      <i class="fa fa-search"></i>
    </span>
    <span class="popup-btn-close">
      <i class="fa fa-times-circle"></i>
    </span>
    <div class="local-search-input-wrapper">
      <input autocomplete="off" placeholder="搜索..." spellcheck="false" type="text" id="local-search-input">
    </div>
  </div>
  <div id="local-search-result"></div>
</div>


    </div>

</nav>


</div>
    </header>


    <main id="main" class="main">
      <div class="main-inner">
        <div class="content-wrap">


          <div id="content" class="content">


  <div id="posts" class="posts-expand">


    <div class="reading-progress-bar"></div>


  <article class="post post-type-normal" itemscope="" itemtype="http://schema.org/Article">


  <div class="post-block">
    <link itemprop="mainEntityOfPage" href="https://www.dudefu.tk/Spark项目落地实战以及日常大数据开发注意事项.html">

    <span hidden itemprop="author" itemscope="" itemtype="http://schema.org/Person">
      <meta itemprop="name" content="Daniel X">
      <meta itemprop="description" content="專注于大数据技術，分享干货">
      <meta itemprop="image" content="https://hexoblog-1254111960.cos.ap-guangzhou.myqcloud.com/HexoBlog-tou.jpg">
    </span>

    <span hidden itemprop="publisher" itemscope="" itemtype="http://schema.org/Organization">
      <meta itemprop="name" content="The Future">
    </span>


      <header class="post-header">


          <h2 class="post-title" itemprop="name headline">Spark项目落地实战以及日常大数据开发注意事项


          </h2>


        <div class="post-meta">
          <span class="post-time">


              <span class="post-meta-item-icon">
                <i class="fa fa-calendar-o"></i>
              </span>

                <span class="post-meta-item-text">发表于</span>


              <time title="创建时间：2021-01-12 12:06:11 / 修改时间：12:08:55" itemprop="dateCreated datePublished" datetime="2021-01-12T12:06:11+08:00">2021-01-12</time>


          </span>


            <span class="post-category">

              <span class="post-meta-divider">|</span>

              <span class="post-meta-item-icon">
                <i class="fa fa-folder-o"></i>
              </span>

                <span class="post-meta-item-text">分类于</span>


                <span itemprop="about" itemscope="" itemtype="http://schema.org/Thing"><a href="/categories/大数据/" itemprop="url" rel="index"><span itemprop="name">大数据</span></a></span>


            </span>


              <span class="post-comments-count">
                <span class="post-meta-divider">|</span>
                <span class="post-meta-item-icon">
                  <i class="fa fa-comment-o"></i>
                </span>

                <span class="post-meta-item-text">评论数：</span>
                <a href="/Spark项目落地实战以及日常大数据开发注意事项.html#comments" itemprop="discussionUrl">
                  <span class="post-comments-count valine-comment-count" data-xid="/Spark项目落地实战以及日常大数据开发注意事项.html" itemprop="commentCount"></span>
                </a>
              </span>


            <span id="/Spark项目落地实战以及日常大数据开发注意事项.html" class="leancloud_visitors" data-flag-title="Spark项目落地实战以及日常大数据开发注意事项">
              <span class="post-meta-divider">|</span>
              <span class="post-meta-item-icon">
                <i class="fa fa-eye"></i>
              </span>

                <span class="post-meta-item-text">阅读次数：</span>

                <span class="leancloud-visitors-count"></span>
            </span>


        </div>
      </header>


    <div class="post-body" itemprop="articleBody">


        <p>Spark简介 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7g9ip7j30fa0770tg.jpg" alt="img"></p>
<a id="more"></a>
<p>基于内存的分布式集群计算平台</p>
<p>可适配 Python、Java、Scala、SQL</p>
<p>拓展功能：机器学习、流式计算、图计算</p>
<p>Spark特点 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7iy2qnj30fa08nju4.jpg" alt="img"></p>
<p>高效 </p>
<ul>
<li>内存计算引擎</li>
<li>DAG图</li>
<li>比MapReduce快10～100倍</li>
</ul>
<h4 id="易用"><a href="#易用" class="headerlink" title="易用"></a>易用</h4><ul>
<li>提供丰富的API，支持Java，Scala， Python</li>
<li>代码量小</li>
</ul>
<h4 id="与Hadoop集成"><a href="#与Hadoop集成" class="headerlink" title="与Hadoop集成"></a>与Hadoop集成</h4><ul>
<li>读写HDFS、Hbase、Hive</li>
<li>和Yarn集成</li>
</ul>
<p>与Oracle存过的对比 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7lcf7aj30fa03sgmk.jpg" alt="img"></p>
<p>Spark应用场景 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7p5ynzj30fa078glw.jpg" alt="img"></p>
<ul>
<li>数据仓库</li>
<li>机器学习</li>
<li>海量数据离线分析</li>
<li>实时数据流处理</li>
</ul>
<p>基本概念 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7scd41j30fa07640c.jpg" alt="img"></p>
<p>集群架构 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7vircej30fa09a750.jpg" alt="img"></p>
<ul>
<li>集群资源管理器（Cluster Manager）</li>
<li>运行作业任务的工作节点（Worker Node）</li>
<li>每个应用的任务控制节点（Driver）</li>
<li>每个工作节点上负责具体任务的执行进程 （Executor）</li>
<li>资源管理器Mesos或YARN</li>
</ul>
<p>任务执行流程 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks7ygkecj30fa09sta6.jpg" alt="img"></p>
<ol>
<li>首先为应用构建起基本的运行环境，即由 Driver创建一个SparkContext，进行资源 的申请、任务的分配和监控</li>
<li>资源管理器为Executor分配资源，并启动 Executor进程</li>
<li>SparkContext根据RDD的依赖关系构建 DAG图，DAG图提交给DAGScheduler解 析成Stage，然后把一个个TaskSet提交给 底层调度器TaskScheduler处理； Executor向SparkContext申请Task，Task Scheduler将Task发放给Executor运行， 并提供应用程序代码</li>
<li>Task在Executor上运行，把执行结果反馈 给TaskScheduler，然后反馈给 DAGScheduler，运行完毕后写入数据并 释放所有资源。</li>
</ol>
<p>数据处理过程 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks81s5qnj30fa05574q.jpg" alt="img"></p>
<ol>
<li>读入外部数据源</li>
<li>转换算子进行数据处理</li>
<li>动作算子进行处理流程触发</li>
<li>处理完成输出结果</li>
</ol>
<p>常用算子-转换 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks842uowj30fa07bq5d.jpg" alt="img"></p>
<p>开发案例–集团电信三码低效资产分析 </p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks86w7u9j30fa08cac6.jpg" alt="img"></p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks89mfdaj30fa08k40z.jpg" alt="img"></p>
<p>Spark很香、也很坑</p>
<p>坑1：无法自定义自增序列</p>
<p>坑2：Spark Stage之间的血缘冗长</p>
<p>坑3：直连Oracle读取慢</p>
<p>坑4：时间格式支持不友好</p>
<p>常见问题1-无法自定义自增序列 </p>
<p>问题阐述：</p>
<p>在不同的业务逻辑中，由于会存在多种维度的分析，但是他们的结果是写入到同一张表格中的。在oracle中执行的时候是根据oracle中定义的序列来保证ID的唯一性，但是 我们代码实现的时候采用的数据加载模式时无法加载oracle中的序列，并且读取序列也会收到oracle序列缓冲的影响。所以在业务逻辑处理上我们得自己定义一个属于我们 业务的ID序列，并且需要保证唯一性。</p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8dail0j30fa0csjvb.jpg" alt="img"></p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8g81p6j30fa0chq75.jpg" alt="img"></p>
<p>常见问题2-血缘关系冗长 </p>
<p>问题阐述：</p>
<p>由于SparkSQL在解析成ATS树时会向上追溯血缘并重复解析，且随着血缘关系的增长ATS树会变的越来越复杂，导致任务执行效率会严重降低。具体表象为 Spark任务在执行过程中会卡住不动，程序继续卡顿几个小时之后才会开始继续运行。</p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8j994gj30fa07rta1.jpg" alt="img"></p>
<p>方案一：checkpoints方式切割方式</p>
<p>方案二：hdfs落地，使用时二次读取</p>
<p>常见问题3-读Oracle速率慢 </p>
<p>问题阐述：</p>
<p>在读取Oracle时，数据表未做分区，程序无法通过指定分区并行加载数据，且为了减小数据库IO压力，采用限制高频、数据读取限制等策略，导致读取Oracle速 率很慢，影响计算效率。</p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8lykeaj30fa0enwkb.jpg" alt="img"></p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8p5bh0j30fa0cc78s.jpg" alt="img"></p>
<p>常见问题4-Oracle时间格式支持不友好 </p>
<p>问题阐述：</p>
<p>park在读写Oracle时date类型数据容易丢失精度，例如： Oracle中 2019-12-20 05:44:30读取后为2019-12-20， Spark中2019-12-20 05:44:30写入后变成2019-12-20 00:00:00</p>
<p><img src="https://tva1.sinaimg.cn/large/008eGmZEgy1gmks8sw7yuj30fa0bc456.jpg" alt="img"></p>
<p>解决方案：</p>
<p>Oracle方言,即自定义一种数据库解释语言，实际上的实现 为数据的类型转换。OracleDateTypeInit.oracleInit()</p>


    </div>


    <footer class="post-footer">


        <div class="post-tags">

            <a href="/tags/spark/" rel="tag"><i class="fa fa-tag"></i> spark</a>

        </div>


        <div class="post-widgets">


          <div class="social_share">

              <div>


<script src="//cdn.jsdelivr.net/npm/ilyabirman-likely@2/release/likely.js"></script>


<link rel="stylesheet" href="//cdn.jsdelivr.net/npm/ilyabirman-likely@2/release/likely.css">


<div class="likely likely-light">

 	 	<div class="twitter">Tweet</div>

 	 	<div class="facebook">Share</div>

 	 	<div class="linkedin">Link</div>

 	 	<div class="gplus">Plus</div>

 	 	<div class="vkontakte">Share</div>

 	 	<div class="odnoklassniki">Class</div>

 	 	<div class="telegram">Send</div>

 	 	<div class="whatsapp">Send</div>

 	 	<div class="pinterest">Pin</div>

</div>

              </div>


              <div>

  <div class="bdsharebuttonbox">
    <a href="#" class="bds_tsina" data-cmd="tsina" title="分享到新浪微博"></a>
    <a href="#" class="bds_douban" data-cmd="douban" title="分享到豆瓣网"></a>
    <a href="#" class="bds_sqq" data-cmd="sqq" title="分享到QQ好友"></a>
    <a href="#" class="bds_qzone" data-cmd="qzone" title="分享到QQ空间"></a>
    <a href="#" class="bds_weixin" data-cmd="weixin" title="分享到微信"></a>
    <a href="#" class="bds_tieba" data-cmd="tieba" title="分享到百度贴吧"></a>
    <a href="#" class="bds_twi" data-cmd="twi" title="分享到Twitter"></a>
    <a href="#" class="bds_fbook" data-cmd="fbook" title="分享到Facebook"></a>
    <a href="#" class="bds_more" data-cmd="more"></a>
    <a class="bds_count" data-cmd="count"></a>
  </div>
  <script>
    window._bd_share_config = {
      "common": {
        "bdText": "",
        "bdMini": "2",
        "bdMiniList": false,
        "bdPic": ""
      },
      "share": {
        "bdSize": "16",
        "bdStyle": "0"
      },
      "image": {
        "viewList": ["tsina", "douban", "sqq", "qzone", "weixin", "twi", "fbook"],
        "viewText": "分享到：",
        "viewSize": "16"
      }
    }
  </script>

<script>
  with(document)0[(getElementsByTagName('head')[0]||body).appendChild(createElement('script')).src='//bdimg.share.baidu.com/static/api/js/share.js?cdnversion='+~(-new Date()/36e5)];
</script>

              </div>

          </div>

        </div>


        <div class="post-nav">
          <div class="post-nav-next post-nav-item">

              <a href="/Elasticsearch查询速度为什么这么快？.html" rel="next" title="Elasticsearch查询速度为什么这么快？">
                <i class="fa fa-chevron-left"></i> Elasticsearch查询速度为什么这么快？
              </a>

          </div>

          <span class="post-nav-divider"></span>

          <div class="post-nav-prev post-nav-item">

              <a href="/基于云计算和大数据的模拟车辆行车监控系统.html" rel="prev" title="基于云计算和大数据的模拟车辆行车监控系统">
                基于云计算和大数据的模拟车辆行车监控系统 <i class="fa fa-chevron-right"></i>
              </a>

          </div>
        </div>


    </footer>
  </div>


  </article>


  </div>


          </div>


    <div class="comments" id="comments">
      <div id="lv-container" data-id="city" data-uid="MTAyMC8yOTk3My82NTM4"></div>
    </div>


        </div>


  <div class="sidebar-toggle">
    <div class="sidebar-toggle-line-wrap">
      <span class="sidebar-toggle-line sidebar-toggle-line-first"></span>
      <span class="sidebar-toggle-line sidebar-toggle-line-middle"></span>
      <span class="sidebar-toggle-line sidebar-toggle-line-last"></span>
    </div>
  </div>

  <aside id="sidebar" class="sidebar">
    <div class="sidebar-inner">


        <ul class="sidebar-nav motion-element">
          <li class="sidebar-nav-toc sidebar-nav-active" data-target="post-toc-wrap">
            文章目录
          </li>
          <li class="sidebar-nav-overview" data-target="site-overview-wrap">
            站点概览
          </li>
        </ul>


      <div class="site-overview-wrap sidebar-panel">
        <div class="site-overview">
          <div class="site-author motion-element" itemprop="author" itemscope="" itemtype="http://schema.org/Person">

              <img class="site-author-image" itemprop="image" src="https://hexoblog-1254111960.cos.ap-guangzhou.myqcloud.com/HexoBlog-tou.jpg" alt="Daniel X">

              <p class="site-author-name" itemprop="name">Daniel X</p>
              <div class="site-description motion-element" itemprop="description">專注于大数据技術，分享干货</div>
          </div>


            <nav class="site-state motion-element">

                <div class="site-state-item site-state-posts">

                  <a href="/archives/">

                    <span class="site-state-item-count">125</span>
                    <span class="site-state-item-name">日志</span>
                  </a>
                </div>


                <div class="site-state-item site-state-categories">


                      <a href="/categories">


                    <span class="site-state-item-count">15</span>
                    <span class="site-state-item-name">分类</span>
                  </a>
                </div>


                <div class="site-state-item site-state-tags">


                      <a href="/tags">


                    <span class="site-state-item-count">63</span>
                    <span class="site-state-item-name">标签</span>
                  </a>
                </div>

            </nav>


            <div class="links-of-author motion-element">

                <span class="links-of-author-item">


                  <span class="exturl" data-url="aHR0cHM6Ly9naXRodWIuY29tL2R1ZGVmdQ==" title="GitHub &rarr; https://github.com/dudefu"><i class="fa fa-fw fa-github"></i>GitHub</span>
                </span>

                <span class="links-of-author-item">


                  <span class="exturl" data-url="bWFpbHRvOmR1ZGVmdUBmb3htYWlsLmNvbT9zdWJqZWN0PUhlbGxvJTIwYWdhaW4=" title="E-mail &rarr; mailto:dudefu@foxmail.com?subject=Hello%20again"><i class="fa fa-fw fa-envelope"></i>E-mail</span>
                </span>

                <span class="links-of-author-item">


                  <span class="exturl" data-url="aHR0cHM6Ly93ZWliby5jb20vZHVkZWZ1" title="Weibo &rarr; https://weibo.com/dudefu"><i class="fa fa-fw fa-weibo"></i>Weibo</span>
                </span>

                <span class="links-of-author-item">


                  <span class="exturl" data-url="aHR0cHM6Ly93cGEucXEuY29tL21zZ3JkP3Y9MyZ1aW49MTU3NzU3MTk1OSZzaXRlPWR1ZGVmdS5pbmZvJm1lbnU9eWVz" title="QQ &rarr; https://wpa.qq.com/msgrd?v=3&uin=1577571959&site=dudefu.info&menu=yes"><i class="fa fa-fw fa-qq"></i>QQ</span>
                </span>

            </div>


        </div>
      </div>


      <!--noindex-->
        <div class="post-toc-wrap motion-element sidebar-panel sidebar-panel-active">
          <div class="post-toc">


              <div class="post-toc-content"><ol class="nav"><li class="nav-item nav-level-4"><a class="nav-link" href="#易用"><span class="nav-number">1.</span> <span class="nav-text">易用</span></a></li><li class="nav-item nav-level-4"><a class="nav-link" href="#与Hadoop集成"><span class="nav-number">2.</span> <span class="nav-text">与Hadoop集成</span></a></li></ol></div>


          </div>
        </div>
      <!--/noindex-->


    </div>
  </aside>


      </div>
    </main>

    <footer id="footer" class="footer">
      <div class="footer-inner">
        <div class="copyright">  <span class="exturl" data-url="aHR0cDovL3d3dy5iZWlhbi5taWl0Lmdvdi5jbg==">粤ICP备18110871号 </span>&copy; 2017 – <span itemprop="copyrightYear">2021</span>
  <span class="with-love" id="animate">
    <i class="fa fa-spinner"></i>
  </span>
  <span class="author" itemprop="copyrightHolder">dudefu</span>


</div>

<!--

  <div class="powered-by">由 <span class="exturl theme-link" data-url="aHR0cHM6Ly9oZXhvLmlv">Hexo</span> 强力驱动 v3.8.0</div>


  <span class="post-meta-divider">|</span>


  <div class="theme-info">主题 – <span class="exturl theme-link" data-url="aHR0cHM6Ly90aGVtZS1uZXh0Lm9yZw==">NexT.Pisces</span> v7.1.2</div>

-->


      </div>
    </footer>


      <div class="back-to-top">
        <i class="fa fa-arrow-up"></i>

          <span id="scrollpercent"><span>0</span>%</span>

      </div>


  </div>


<script>
  if (Object.prototype.toString.call(window.Promise) !== '[object Function]') {
    window.Promise = null;
  }
</script>


  <script color="26,26,26" opacity="0.5" zindex="-1" count="99" src="//cdn.jsdelivr.net/gh/theme-next/theme-next-canvas-nest@1/canvas-nest.min.js"></script>


  <script id="ribbon" size="300" alpha="0.6" zindex="-1" src="/lib/canvas-ribbon/canvas-ribbon.js"></script>


  <script src="/lib/jquery/index.js?v=3.4.1"></script>


  <script src="/lib/velocity/velocity.min.js?v=1.2.1"></script>


  <script src="/lib/velocity/velocity.ui.min.js?v=1.2.1"></script>


  <script src="/lib/reading_progress/reading_progress.js"></script>


  <script src="/js/utils.js?v=7.1.2"></script>

  <script src="/js/motion.js?v=7.1.2"></script>


  <script src="/js/affix.js?v=7.1.2"></script>

  <script src="/js/schemes/pisces.js?v=7.1.2"></script>


  <script src="/js/scrollspy.js?v=7.1.2"></script>
<script src="/js/post-details.js?v=7.1.2"></script>


  <script src="/js/next-boot.js?v=7.1.2"></script>


  <script src="/js/js.cookie.js?v=7.1.2"></script>
  <script src="/js/scroll-cookie.js?v=7.1.2"></script>


  <script src="/js/exturl.js?v=7.1.2"></script>


<script src="//cdn1.lncld.net/static/js/3.11.1/av-min.js"></script>


<script src="//unpkg.com/valine/dist/Valine.min.js"></script>

<script>
  var GUEST = ['nick', 'mail', 'link'];
  var guest = 'nick,mail,link';
  guest = guest.split(',').filter(function(item) {
    return GUEST.indexOf(item) > -1;
  });
  new Valine({
    el: '#comments',
    verify: true,
    notify: true,
    appId: '1N5rpk874DGudJw2wCL9J011-gzGzoHsz',
    appKey: '9Y83e6suJgx567wtxhKy45IN',
    placeholder: 'Just go go',
    avatar: 'mm',
    meta: guest,
    pageSize: '10' || 10,
    visitor: true,
    lang: 'zk-cn' || 'zh-cn'
  });
</script>


    <script>
  window.livereOptions = {
    refer: 'Spark项目落地实战以及日常大数据开发注意事项.html'
  };
  (function(d, s) {
    var j, e = d.getElementsByTagName(s)[0];
    if (typeof LivereTower === 'function') { return; }
    j = d.createElement(s);
    j.src = 'https://cdn-city.livere.com/js/embed.dist.js';
    j.async = true;
    e.parentNode.insertBefore(j, e);
  })(document, 'script');
</script>


  <script>
    // Popup Window;
    var isfetched = false;
    var isXml = true;
    // Search DB path;
    var search_path = "search.xml";
    if (search_path.length === 0) {
      search_path = "search.xml";
    } else if (/json$/i.test(search_path)) {
      isXml = false;
    }
    var path = "/" + search_path;
    // monitor main search box;

    var onPopupClose = function (e) {
      $('.popup').hide();
      $('#local-search-input').val('');
      $('.search-result-list').remove();
      $('#no-result').remove();
      $(".local-search-pop-overlay").remove();
      $('body').css('overflow', '');
    }

    function proceedsearch() {
      $("body")
        .append('<div class="search-popup-overlay local-search-pop-overlay"></div>')
        .css('overflow', 'hidden');
      $('.search-popup-overlay').click(onPopupClose);
      $('.popup').toggle();
      var $localSearchInput = $('#local-search-input');
      $localSearchInput.attr("autocapitalize", "none");
      $localSearchInput.attr("autocorrect", "off");
      $localSearchInput.focus();
    }

    // search function;
    var searchFunc = function(path, search_id, content_id) {
      'use strict';

      // start loading animation
      $("body")
        .append('<div class="search-popup-overlay local-search-pop-overlay">' +
          '<div id="search-loading-icon">' +
          '<i class="fa fa-spinner fa-pulse fa-5x fa-fw"></i>' +
          '</div>' +
          '</div>')
        .css('overflow', 'hidden');
      $("#search-loading-icon").css('margin', '20% auto 0 auto').css('text-align', 'center');


      $.ajax({
        url: path,
        dataType: isXml ? "xml" : "json",
        async: true,
        success: function(res) {
          // get the contents from search data
          isfetched = true;
          $('.popup').detach().appendTo('.header-inner');
          var datas = isXml ? $("entry", res).map(function() {
            return {
              title: $("title", this).text(),
              content: $("content",this).text(),
              url: $("url" , this).text()
            };
          }).get() : res;
          var input = document.getElementById(search_id);
          var resultContent = document.getElementById(content_id);
          var inputEventFunction = function() {
            var searchText = input.value.trim().toLowerCase();
            var keywords = searchText.split(/[\s\-]+/);
            if (keywords.length > 1) {
              keywords.push(searchText);
            }
            var resultItems = [];
            if (searchText.length > 0) {
              // perform local searching
              datas.forEach(function(data) {
                var isMatch = false;
                var hitCount = 0;
                var searchTextCount = 0;
                var title = data.title.trim();
                var titleInLowerCase = title.toLowerCase();
                var content = data.content.trim().replace(/<[^>]+>/g,"");

                var contentInLowerCase = content.toLowerCase();
                var articleUrl = decodeURIComponent(data.url).replace(/\/{2,}/g, '/');
                var indexOfTitle = [];
                var indexOfContent = [];
                // only match articles with not empty titles
                if(title != '') {
                  keywords.forEach(function(keyword) {
                    function getIndexByWord(word, text, caseSensitive) {
                      var wordLen = word.length;
                      if (wordLen === 0) {
                        return [];
                      }
                      var startPosition = 0, position = [], index = [];
                      if (!caseSensitive) {
                        text = text.toLowerCase();
                        word = word.toLowerCase();
                      }
                      while ((position = text.indexOf(word, startPosition)) > -1) {
                        index.push({position: position, word: word});
                        startPosition = position + wordLen;
                      }
                      return index;
                    }

                    indexOfTitle = indexOfTitle.concat(getIndexByWord(keyword, titleInLowerCase, false));
                    indexOfContent = indexOfContent.concat(getIndexByWord(keyword, contentInLowerCase, false));
                  });
                  if (indexOfTitle.length > 0 || indexOfContent.length > 0) {
                    isMatch = true;
                    hitCount = indexOfTitle.length + indexOfContent.length;
                  }
                }

                // show search results

                if (isMatch) {
                  // sort index by position of keyword

                  [indexOfTitle, indexOfContent].forEach(function (index) {
                    index.sort(function (itemLeft, itemRight) {
                      if (itemRight.position !== itemLeft.position) {
                        return itemRight.position - itemLeft.position;
                      } else {
                        return itemLeft.word.length - itemRight.word.length;
                      }
                    });
                  });

                  // merge hits into slices

                  function mergeIntoSlice(text, start, end, index) {
                    var item = index[index.length - 1];
                    var position = item.position;
                    var word = item.word;
                    var hits = [];
                    var searchTextCountInSlice = 0;
                    while (position + word.length <= end && index.length != 0) {
                      if (word === searchText) {
                        searchTextCountInSlice++;
                      }
                      hits.push({position: position, length: word.length});
                      var wordEnd = position + word.length;

                      // move to next position of hit

                      index.pop();
                      while (index.length != 0) {
                        item = index[index.length - 1];
                        position = item.position;
                        word = item.word;
                        if (wordEnd > position) {
                          index.pop();
                        } else {
                          break;
                        }
                      }
                    }
                    searchTextCount += searchTextCountInSlice;
                    return {
                      hits: hits,
                      start: start,
                      end: end,
                      searchTextCount: searchTextCountInSlice
                    };
                  }

                  var slicesOfTitle = [];
                  if (indexOfTitle.length != 0) {
                    slicesOfTitle.push(mergeIntoSlice(title, 0, title.length, indexOfTitle));
                  }

                  var slicesOfContent = [];
                  while (indexOfContent.length != 0) {
                    var item = indexOfContent[indexOfContent.length - 1];
                    var position = item.position;
                    var word = item.word;
                    // cut out 100 characters
                    var start = position - 20;
                    var end = position + 80;
                    if(start < 0){
                      start = 0;
                    }
                    if (end < position + word.length) {
                      end = position + word.length;
                    }
                    if(end > content.length){
                      end = content.length;
                    }
                    slicesOfContent.push(mergeIntoSlice(content, start, end, indexOfContent));
                  }

                  // sort slices in content by search text's count and hits' count

                  slicesOfContent.sort(function (sliceLeft, sliceRight) {
                    if (sliceLeft.searchTextCount !== sliceRight.searchTextCount) {
                      return sliceRight.searchTextCount - sliceLeft.searchTextCount;
                    } else if (sliceLeft.hits.length !== sliceRight.hits.length) {
                      return sliceRight.hits.length - sliceLeft.hits.length;
                    } else {
                      return sliceLeft.start - sliceRight.start;
                    }
                  });

                  // select top N slices in content

                  var upperBound = parseInt('1');
                  if (upperBound >= 0) {
                    slicesOfContent = slicesOfContent.slice(0, upperBound);
                  }

                  // highlight title and content

                  function highlightKeyword(text, slice) {
                    var result = '';
                    var prevEnd = slice.start;
                    slice.hits.forEach(function (hit) {
                      result += text.substring(prevEnd, hit.position);
                      var end = hit.position + hit.length;
                      result += '<b class="search-keyword">' + text.substring(hit.position, end) + '</b>';
                      prevEnd = end;
                    });
                    result += text.substring(prevEnd, slice.end);
                    return result;
                  }

                  var resultItem = '';

                  if (slicesOfTitle.length != 0) {
                    resultItem += "<li><a href='" + articleUrl + "' class='search-result-title'>" + highlightKeyword(title, slicesOfTitle[0]) + "</a>";
                  } else {
                    resultItem += "<li><a href='" + articleUrl + "' class='search-result-title'>" + title + "</a>";
                  }

                  slicesOfContent.forEach(function (slice) {
                    resultItem += "<a href='" + articleUrl + "'>" +
                      "<p class=\"search-result\">" + highlightKeyword(content, slice) +
                      "...</p>" + "</a>";
                  });

                  resultItem += "</li>";
                  resultItems.push({
                    item: resultItem,
                    searchTextCount: searchTextCount,
                    hitCount: hitCount,
                    id: resultItems.length
                  });
                }
              })
            };
            if (keywords.length === 1 && keywords[0] === "") {
              resultContent.innerHTML = '<div id="no-result"><i class="fa fa-search fa-5x"></i></div>'
            } else if (resultItems.length === 0) {
              resultContent.innerHTML = '<div id="no-result"><i class="fa fa-frown-o fa-5x"></i></div>'
            } else {
              resultItems.sort(function (resultLeft, resultRight) {
                if (resultLeft.searchTextCount !== resultRight.searchTextCount) {
                  return resultRight.searchTextCount - resultLeft.searchTextCount;
                } else if (resultLeft.hitCount !== resultRight.hitCount) {
                  return resultRight.hitCount - resultLeft.hitCount;
                } else {
                  return resultRight.id - resultLeft.id;
                }
              });
              var searchResultList = '<ul class=\"search-result-list\">';
              resultItems.forEach(function (result) {
                searchResultList += result.item;
              })
              searchResultList += "</ul>";
              resultContent.innerHTML = searchResultList;
            }
          }

          if ('auto' === 'auto') {
            input.addEventListener('input', inputEventFunction);
          } else {
            $('.search-icon').click(inputEventFunction);
            input.addEventListener('keypress', function (event) {
              if (event.keyCode === 13) {
                inputEventFunction();
              }
            });
          }

          // remove loading animation
          $(".local-search-pop-overlay").remove();
          $('body').css('overflow', '');

          proceedsearch();
        }
      });
    }

    // handle and trigger popup window;
    $('.popup-trigger').click(function(e) {
      e.stopPropagation();
      if (isfetched === false) {
        searchFunc(path, 'local-search-input', 'local-search-result');
      } else {
        proceedsearch();
      };
    });

    $('.popup-btn-close').click(onPopupClose);
    $('.popup').click(function(e){
      e.stopPropagation();
    });
    $(document).on('keyup', function (event) {
      var shouldDismissSearchPopup = event.which === 27 &&
        $('.search-popup').is(':visible');
      if (shouldDismissSearchPopup) {
        onPopupClose();
      }
    });
  </script>


  <script src="https://www.gstatic.com/firebasejs/4.6.0/firebase.js"></script>
  <script src="https://www.gstatic.com/firebasejs/4.6.0/firebase-firestore.js"></script>

    <script src="https://cdnjs.cloudflare.com/ajax/libs/bluebird/3.5.1/bluebird.core.min.js"></script>

  <script>
    (function () {

      firebase.initializeApp({
        apiKey: '',
        projectId: ''
      })

      function getCount(doc, increaseCount) {
        //increaseCount will be false when not in article page

        return doc.get().then(function (d) {
          var count
          if (!d.exists) { //has no data, initialize count
            if (increaseCount) {
              doc.set({
                count: 1
              })
              count = 1
            }
            else {
              count = 0
            }
          }
          else { //has data
            count = d.data().count
            if (increaseCount) {
              if (!(window.localStorage && window.localStorage.getItem(title))) { //if first view this article
                doc.set({ //increase count
                  count: count + 1
                })
                count++
              }
            }
          }
          if (window.localStorage && increaseCount) { //mark as visited
            localStorage.setItem(title, true)
          }

          return count
        })
      }

      function appendCountTo(el) {
        return function (count) {
          $(el).append(
            $('<span>').addClass('post-visitors-count').append(
              $('<span>').addClass('post-meta-divider').text('|')
            ).append(
              $('<span>').addClass('post-meta-item-icon').append(
                $('<i>').addClass('fa fa-users')
              )
              ).append($('<span>').text('阅读次数 ' + count))
          )
        }
      }

      var db = firebase.firestore()
      var articles = db.collection('articles')

      //https://hexo.io/docs/variables.html
      var isPost = 'Spark项目落地实战以及日常大数据开发注意事项'.length > 0
      var isArchive = '' === 'true'
      var isCategory = ''.length > 0
      var isTag = ''.length > 0

      if (isPost) { //is article page
        var title = 'Spark项目落地实战以及日常大数据开发注意事项'
        var doc = articles.doc(title)

        getCount(doc, true).then(appendCountTo($('.post-meta')))
      }
      else if (!isArchive && !isCategory && !isTag) { //is index page
        var titles = [] //array to titles

        var postsstr = '' //if you have a better way to get titles of posts, please change it
        eval(postsstr)

        var promises = titles.map(function (title) {
          return articles.doc(title)
        }).map(function (doc) {
          return getCount(doc)
        })
        Promise.all(promises).then(function (counts) {
          var metas = $('.post-meta')
          counts.forEach(function (val, idx) {
            appendCountTo(metas[idx])(val)
          })
        })
      }
    })()
  </script>


<script>
if ($('body').find('div.pdf').length) {
  $.ajax({
    type: 'GET',
    url: '//cdn.jsdelivr.net/npm/pdfobject@2/pdfobject.min.js',
    dataType: 'script',
    cache: true,
    success: function() {
      $('body').find('div.pdf').each(function(i, o) {
        PDFObject.embed($(o).attr('target'), $(o), {
          pdfOpenParams: {
            navpanes: 0,
            toolbar: 0,
            statusbar: 0,
            pagemode: 'thumbs',
            view: 'FitH'
          },
          PDFJS_URL: '/lib/pdf/web/viewer.html',
          height: $(o).attr('height') || '500px'
        });
      });
    },
  });
}
</script>


<script>
if ($('body').find('pre.mermaid').length) {
  $.ajax({
    type: 'GET',
    url: '//cdn.jsdelivr.net/npm/mermaid@8/dist/mermaid.min.js',
    dataType: 'script',
    cache: true,
    success: function() {
      mermaid.initialize({
        theme: 'dark',
        logLevel: 3,
        flowchart: { curve: 'linear' },
        gantt: { axisFormat: '%m/%d/%Y' },
        sequence: { actorMargin: 50 }
      });
    }
  });
}
</script>


  <script>
    (function(){
      var bp = document.createElement('script');
      var curProtocol = window.location.protocol.split(':')[0];
      bp.src = (curProtocol === 'https') ? 'https://zz.bdstatic.com/linksubmit/push.js' : 'http://push.zhanzhang.baidu.com/push.js';
      var s = document.getElementsByTagName("script")[0];
      s.parentNode.insertBefore(bp, s);
    })();
  </script>


  <script src="/lib/bookmark/bookmark.min.js?v=1.0"></script>
  <script>

    bookmark.scrollToMark('auto', "#更多");

  </script>


<script>
  $('.highlight').not('.gist .highlight').each(function(i, e) {
    var $wrap = $('<div>').addClass('highlight-wrap');
    $(e).after($wrap);
    $wrap.append($('<button>').addClass('copy-btn').append('复制').on('click', function(e) {
      var code = $(this).parent().find('.code').find('.line').map(function(i, e) {
        return $(e).text();
      }).toArray().join('\n');
      var ta = document.createElement('textarea');
      var yPosition = window.pageYOffset || document.documentElement.scrollTop;
      ta.style.top = yPosition + 'px'; // Prevent page scroll
      ta.style.position = 'absolute';
      ta.style.opacity = '0';
      ta.readOnly = true;
      ta.value = code;
      document.body.appendChild(ta);
      const selection = document.getSelection();
      const selected = selection.rangeCount > 0 ? selection.getRangeAt(0) : false;
      ta.select();
      ta.setSelectionRange(0, code.length);
      ta.readOnly = false;
      var result = document.execCommand('copy');

        if (result) $(this).text('复制成功');
        else $(this).text('复制失败');

      ta.blur(); // For iOS
      $(this).blur();
      if (selected) {
        selection.removeAllRanges();
        selection.addRange(selected);
      }
    })).on('mouseleave', function(e) {
      var $b = $(this).find('.copy-btn');
      setTimeout(function() {
        $b.text('复制');
      }, 300);
    }).append(e);
  })
</script>


</body>
</html>