{"id":1718,"date":"2021-01-31T21:05:18","date_gmt":"2021-01-31T13:05:18","guid":{"rendered":"https:\/\/www.specialwu.com\/?p=1718"},"modified":"2021-04-05T18:03:57","modified_gmt":"2021-04-05T10:03:57","slug":"%e5%a4%a7%e6%95%b0%e6%8d%ae%e8%ae%a1%e7%ae%97%e4%b9%8bmapreduce","status":"publish","type":"post","link":"http:\/\/www.specialwu.com\/?p=1718","title":{"rendered":"\u5927\u6570\u636e\u8ba1\u7b97\u7ec4\u4ef6&#8211;MapReduce"},"content":{"rendered":"<ul>\n<li>\u5b98\u7f51\u7ed9\u51fa\u7684\u4ecb\u7ecd<\/li>\n<\/ul>\n<p>Hadoop MapReduce is a software framework for easily writing applications which process vast amounts of data (multi-terabyte data-sets) in-parallel on large clusters (thousands of nodes) of commodity hardware in a reliable, fault-tolerant manner.<\/p>\n<p>A MapReduce job usually splits the input data-set into independent chunks which are processed by the map tasks in a completely parallel manner. The framework sorts the outputs of the maps, which are then input to the reduce tasks. Typically both the input and the output of the job are stored in a file-system. The framework takes care of scheduling tasks, monitoring them and re-executes the failed tasks.<\/p>\n<p>Typically the compute nodes and the storage nodes are the same, that is, the MapReduce framework and the Hadoop Distributed File System (see HDFS Architecture Guide) are running on the same set of nodes. This configuration allows the framework to effectively schedule tasks on the nodes where data is already present, resulting in very high aggregate bandwidth across the cluster.<\/p>\n<p>The MapReduce framework consists of a single master ResourceManager, one worker NodeManager per cluster-node, and MRAppMaster per application (see YARN Architecture Guide).<\/p>\n<p>Minimally, applications specify the input\/output locations and supply map and reduce functions via implementations of appropriate interfaces and\/or abstract-classes. These, and other job parameters, comprise the job configuration.<\/p>\n<p>The Hadoop job client then submits the job (jar\/executable etc.) and configuration to the ResourceManager which then assumes the responsibility of distributing the software\/configuration to the workers, scheduling tasks and monitoring them, providing status and diagnostic information to the job-client.<\/p>\n<p>Although the Hadoop framework is implemented in Java\u2122, MapReduce applications need not be written in Java.<\/p>\n<p>Hadoop Streaming is a utility which allows users to create and run jobs with any executables (e.g. shell utilities) as the mapper and\/or the reducer.<\/p>\n<p>Hadoop Pipes is a SWIG-compatible C++ API to implement MapReduce applications (non JNI\u2122 based).<\/p>\n<ul>\n<li>Inputs and Outputs<\/li>\n<\/ul>\n<p>The MapReduce framework operates exclusively on &lt;key, value> pairs, that is, the framework views the input to the job as a set of &lt;key, value> pairs and produces a set of &lt;key, value> pairs as the output of the job, conceivably of different types.<\/p>\n<p>The key and value classes have to be serializable by the framework and hence need to implement the Writable interface. Additionally, the key classes have to implement the WritableComparable interface to facilitate sorting by the framework.<\/p>\n<p>Input and Output types of a MapReduce job:<\/p>\n<p>(input) &lt;k1, v1> -> map -> &lt;k2, v2> -> combine -> &lt;k2, v2> -> reduce -> &lt;k3, v3> (output)<\/p>\n<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_61 ez-toc-wrap-center counter-hierarchy ez-toc-counter ez-toc-custom ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title \" >\u76ee\u5f55<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #000000;color:#000000\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #000000;color:#000000\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E5%89%8D%E8%A8%80\" title=\"\u524d\u8a00\">\u524d\u8a00<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E4%B8%80%E3%80%81MapReduce%E6%98%AF%E4%BB%80%E4%B9%88\" title=\"\u4e00\u3001MapReduce\u662f\u4ec0\u4e48\">\u4e00\u3001MapReduce\u662f\u4ec0\u4e48<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E4%BA%8C%E3%80%81MapReduce%E8%AF%A5%E6%80%8E%E4%B9%88%E7%94%A8\" title=\"\u4e8c\u3001MapReduce\u8be5\u600e\u4e48\u7528\">\u4e8c\u3001MapReduce\u8be5\u600e\u4e48\u7528<\/a><ul class='ez-toc-list-level-2' ><li class='ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#MapReduce%E9%85%8D%E7%BD%AE\" title=\"MapReduce\u914d\u7f6e\">MapReduce\u914d\u7f6e<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#MapReduce%E7%9A%84%E4%BD%BF%E7%94%A8\" title=\"MapReduce\u7684\u4f7f\u7528\">MapReduce\u7684\u4f7f\u7528<\/a><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E4%B8%89%E3%80%81MapReduce%E7%9A%84%E6%A0%B8%E5%BF%83%E9%97%AE%E9%A2%98\" title=\"\u4e09\u3001MapReduce\u7684\u6838\u5fc3\u95ee\u9898\">\u4e09\u3001MapReduce\u7684\u6838\u5fc3\u95ee\u9898<\/a><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><ul class='ez-toc-list-level-3' ><li class='ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#MapReduce%E7%9A%84%E5%87%A0%E7%A7%8D%E6%8E%92%E5%BA%8F%E6%96%B9%E5%BC%8F\" title=\"MapReduce\u7684\u51e0\u79cd\u6392\u5e8f\u65b9\u5f0f\">MapReduce\u7684\u51e0\u79cd\u6392\u5e8f\u65b9\u5f0f<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#MapReduce%E8%B0%83%E4%BC%98\" title=\"MapReduce\u8c03\u4f18\">MapReduce\u8c03\u4f18<\/a><ul class='ez-toc-list-level-4' ><li class='ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#core-sitexml_%E7%9A%84%E8%B0%83%E6%95%B4\" title=\"core-site.xml \u7684\u8c03\u6574\">core-site.xml \u7684\u8c03\u6574<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-10\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#hdfs-sitexml_%E9%85%8D%E7%BD%AE%E6%96%87%E4%BB%B6%E7%9A%84%E8%B0%83%E6%95%B4\" title=\"hdfs-site.xml \u914d\u7f6e\u6587\u4ef6\u7684\u8c03\u6574\">hdfs-site.xml \u914d\u7f6e\u6587\u4ef6\u7684\u8c03\u6574<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-11\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#mapred-sitexml\" title=\"mapred-site.xml\">mapred-site.xml<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-12\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#Yarn-sitexml%E9%85%8D%E7%BD%AE\" title=\"Yarn-site.xml\u914d\u7f6e\">Yarn-site.xml\u914d\u7f6e<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-4'><a class=\"ez-toc-link ez-toc-heading-13\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E9%87%8D%E8%A6%81%E7%9A%84%E5%8F%82%E6%95%B0%E9%85%8D%E7%BD%AE\" title=\"\u91cd\u8981\u7684\u53c2\u6570\u914d\u7f6e\">\u91cd\u8981\u7684\u53c2\u6570\u914d\u7f6e<\/a><\/li><\/ul><\/li><\/ul><\/li><\/ul><\/li><li class='ez-toc-page-1 ez-toc-heading-level-1'><a class=\"ez-toc-link ez-toc-heading-14\" href=\"http:\/\/www.specialwu.com\/?p=1718\/#%E8%A7%A3%E5%86%B3%E4%B8%80%E4%BA%9B%E6%8A%A5%E9%94%99\" title=\"\u89e3\u51b3\u4e00\u4e9b\u62a5\u9519\">\u89e3\u51b3\u4e00\u4e9b\u62a5\u9519<\/a><\/li><\/ul><\/nav><\/div>\n<h1><span class=\"ez-toc-section\" id=\"%E5%89%8D%E8%A8%80\"><\/span>\u524d\u8a00<span class=\"ez-toc-section-end\"><\/span><\/h1>\n<p>MapReduce\u4f5c\u4e3ahadoop\u4e0b\u7684\u4e00\u4e2a\u8ba1\u7b97\u6a21\u5757\uff0c\u90a3\u4e48\u76f8\u5e94\u5177\u5907Hadoop\u7684\u4e00\u4e9b\u7279\u5f81\uff1b\u5728\u65e5\u5e38\u5bf9\u6570\u636e\u5904\u7406\u8fc7\u7a0b\u4e2d\u5e76\u4e0d\u76f4\u63a5\u4f7f\u7528mapreduce\u8fdb\u884c\u6570\u636e\u5904\u7406\uff0c\u76f4\u63a5\u7528mapreduce\u5904\u7406\u6570\u636e\u5e76\u4e0d\u662f\u5f88\u65b9\u4fbf\uff0c\u90a3\u4e48\u5728\u54ea\u91cc\u5e94\u7528MapReduce\u5462\uff1f\u6211\u4eec\u5728\u4f7f\u7528hive SQL\u5904\u7406\u6570\u636e\u65f6\u90a3\u4e48\u5e95\u5c42\u5c31\u4f7f\u7528\u7684\u662fMapReduce\u8fdb\u884c\u6570\u636e\u7684\u8ba1\u7b97\uff0c\u90a3\u4e48\u5c31\u9700\u8981\u4e86\u89e3\u5230MapReduce\u662f\u4ec0\u4e48\uff0c\u4ed6\u662f\u600e\u4e48\u7528\u7684\uff0c\u6211\u4eec\u901a\u5e38\u7528\u6765\u89e3\u51b3\u90a3\u4e9b\u95ee\u9898\u8fd9\u662f\u672c\u7bc7\u5c55\u793a\u7684\u5185\u5bb9\u3002<\/p>\n<h1><span class=\"ez-toc-section\" id=\"%E4%B8%80%E3%80%81MapReduce%E6%98%AF%E4%BB%80%E4%B9%88\"><\/span>\u4e00\u3001MapReduce\u662f\u4ec0\u4e48<span class=\"ez-toc-section-end\"><\/span><\/h1>\n<p><code>Mapreduce\u662f\u4e00\u4e2a\u5206\u5e03\u5f0f\u8fd0\u7b97\u7a0b\u5e8f\u7684\u7f16\u7a0b\u6846\u67b6\uff0c\u662f\u7528\u6237\u5f00\u53d1\u201c\u57fa\u4e8ehadoop\u7684\u6570\u636e\u5206\u6790\u5e94\u7528\u201d\u7684\u6838\u5fc3\u6846\u67b6\uff1b<br \/>\nMapreduce\u6838\u5fc3\u529f\u80fd\u662f\u5c06\u7528\u6237\u7f16\u5199\u7684\u4e1a\u52a1\u903b\u8f91\u4ee3\u7801\u548c\u81ea\u5e26\u9ed8\u8ba4\u7ec4\u4ef6\u6574\u5408\u6210\u4e00\u4e2a\u5b8c\u6574\u7684\u5206\u5e03\u5f0f\u8fd0\u7b97\u7a0b\u5e8f\uff0c\u5e76\u53d1\u8fd0\u884c\u5728\u4e00\u4e2ahadoop\u96c6\u7fa4\u4e0a\u3002<\/code><br \/>\nMapReduce\u4f5c\u4e1a\u901a\u5e38\u5c06\u8f93\u5165\u6570\u636e\u96c6\u5206\u6210\u72ec\u7acb\u7684\u5757\uff0c\u8fd9\u4e9b\u4efb\u52a1\u7531\u5730\u56fe\u4efb\u52a1\u4ee5\u5b8c\u5168\u5e76\u884c\u7684\u65b9\u5f0f\u5904\u7406\u3002\u6846\u67b6\u5bf9\u5730\u56fe\u7684\u8f93\u51fa\u8fdb\u884c\u6392\u5e8f\uff0c\u7136\u540e\u5c06\u5176\u8f93\u5165\u5230reduce\u4efb\u52a1\u3002\u901a\u5e38\uff0c\u4f5c\u4e1a\u7684\u8f93\u5165\u548c\u8f93\u51fa\u90fd\u5b58\u50a8\u5728\u6587\u4ef6\u7cfb\u7edf\u4e2d\u3002\u8be5\u6846\u67b6\u8d1f\u8d23\u5b89\u6392\u4efb\u52a1\uff0c\u76d1\u89c6\u4efb\u52a1\u5e76\u91cd\u65b0\u6267\u884c\u5931\u8d25\u7684\u4efb\u52a1\u3002<br \/>\nMapReduce\u6846\u67b6\u7531\u4e00\u4e2a\u4e3b\u8d44\u6e90\u7ba1\u7406\u5668\uff0c\u6bcf\u4e2a\u7fa4\u96c6\u8282\u70b9\u4e00\u4e2a\u5de5\u4f5c\u5668NodeManager\u548c\u6bcf\u4e2a\u5e94\u7528\u7a0b\u5e8fMRAppMaster\u7ec4\u6210<\/p>\n<h1><span class=\"ez-toc-section\" id=\"%E4%BA%8C%E3%80%81MapReduce%E8%AF%A5%E6%80%8E%E4%B9%88%E7%94%A8\"><\/span>\u4e8c\u3001MapReduce\u8be5\u600e\u4e48\u7528<span class=\"ez-toc-section-end\"><\/span><\/h1>\n<p>\u5728\u7406\u89e3\u600e\u4e48\u7528\u4e4b\u524d\uff0c\u9996\u5148\u660e\u786e\u5b83\u662f\u600e\u4e48\u5b89\u88c5\u914d\u7f6e\u7684<\/p>\n<h2><span class=\"ez-toc-section\" id=\"MapReduce%E9%85%8D%E7%BD%AE\"><\/span>MapReduce\u914d\u7f6e<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>\u7b2c\u4e00\u6b65\uff1a\u5728IDEA\u7684pom.xml\u4e0a\u6dfb\u52a0\u9700\u8981\u5f15\u5165\u7684\u4f9d\u8d56\uff08\u5982\u679c\u4e0d\u5728ideal\u4e0a\u5f00\u53d1\u6216\u8005\u5f00\u53d1\u597d\u4e86\u5ffd\u7565\u6b64\u6b65\u9aa4\uff09<\/p>\n<pre data-language=XML><code class=\"language-markup line-numbers\">&lt;dependencies&gt;\n&lt;dependency&gt;\n            &lt;groupId&gt;org.apache.hadoop&lt;\/groupId&gt;\n            &lt;artifactId&gt;hadoop-client&lt;\/artifactId&gt;\n            &lt;!--    \u6839\u636e\u81ea\u5df1\u7684\u7248\u672c\u8fdb\u884c\u4f9d\u8d56\u5bfc\u5165    --&gt;\n            &lt;version&gt;2.7.7&lt;\/version&gt;\n            &lt;exclusions&gt;\n                &lt;exclusion&gt;\n                    &lt;groupId&gt;com.fasterxml.jackson.core&lt;\/groupId&gt;\n                    &lt;artifactId&gt;jackson-databind&lt;\/artifactId&gt;\n                &lt;\/exclusion&gt;\n            &lt;\/exclusions&gt;\n &lt;\/dependency&gt;\n&lt;\/dependencies&gt;\n<\/code><\/pre>\n<p>\u7b2c\u4e8c\u6b65\uff1aLinux\u4e0aHadoop\u7684\u5b89\u88c5\u6587\u4ef6\u4e0b\u9700\u8981\u5f15\u5165\u7684\u914d\u7f6e<br \/>\n\u96c6\u7fa4\u8fd0\u884c\u6a21\u5f0f\u914d\u7f6e\u5728<code>mapred-site.xml<\/code>\u4e0b\u5f15\u5165\u8be5\u914d\u7f6e<\/p>\n<pre data-language=XML><code class=\"language-markup line-numbers\">&lt;configuration&gt;\n&lt;property&gt;\n   &lt;name&gt;mapreduce.framework.name&lt;\/name&gt;\n   &lt;value&gt;yarn&lt;\/value&gt;\n &lt;\/property&gt;\n&lt;property&gt;\n  &lt;name&gt;jobtracker.thrift.address&lt;\/name&gt;\n  &lt;value&gt;0.0.0.0:9290&lt;\/value&gt;\n&lt;\/property&gt;\n&lt;property&gt;\n  &lt;name&gt;mapred.jobtracker.plugins&lt;\/name&gt;\n  &lt;value&gt;org.apache.hadoop.thriftfs.ThriftJobTrackerPlugin&lt;\/value&gt;\n  &lt;description&gt;Comma-separated list of jobtracker plug-ins to be activated.&lt;\/description&gt;\n&lt;\/property&gt;\n&lt;\/configuration&gt;\n<\/code><\/pre>\n<p>\u7b2c\u4e09\u6b65\uff1a\u6253jar\u5305<\/p>\n<p>ideal\u4e0a\u6253jar\u5305\u8fc7\u7a0b\uff1afile -> project structure -> Artifacts ->add ->jar -> from modules dependencies ->\u9009\u597d\u7c7b\u540eOK<br \/>\n\u9ed8\u8ba4\u8def\u5f84\u662f\uff1aD:\\IdealProject\\bd20201\\out\\artifacts\\strangewugit_jar<br \/>\nBuild -> Build Artifacts -> build<\/p>\n<h2><span class=\"ez-toc-section\" id=\"MapReduce%E7%9A%84%E4%BD%BF%E7%94%A8\"><\/span>MapReduce\u7684\u4f7f\u7528<span class=\"ez-toc-section-end\"><\/span><\/h2>\n<blockquote><p>\n  \u4e0b\u9762\u662f\u5b98\u7f51\u7ed9\u51fa\u7684\u4e00\u4e2a\u5355\u8bcd\u8ba1\u6570\u6848\u4f8b\n<\/p><\/blockquote>\n<pre><code class=\"language-java line-numbers\">import org.apache.hadoop.conf.Configuration;\nimport org.apache.hadoop.fs.FileSystem;\nimport org.apache.hadoop.fs.Path;\nimport org.apache.hadoop.io.LongWritable;\nimport org.apache.hadoop.io.Text;\nimport org.apache.hadoop.mapreduce.Job;\nimport org.apache.hadoop.mapreduce.Mapper;\nimport org.apache.hadoop.mapreduce.Reducer;\nimport org.apache.hadoop.mapreduce.lib.input.FileInputFormat;\nimport org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;\n\nimport java.io.IOException;\n\npublic class MRWordCount {\n\n    public static void main(String[] args) throws Exception {\n\n        \/\/1. \u83b7\u53d6job\u5b9e\u4f8b\n        Configuration conf = new Configuration();\n\/\/        conf.set(\"mapreduce.input.fileinputformat.split.minsize\",\"12313123\");\n        Job job = Job.getInstance(conf);\n\n        \/\/2. \u8bbe\u7f6ejob\u8fd0\u884c\u7684\u4e3b\u7c7b\n        job.setJarByClass(MRWordCount.class);\n\n        \/\/3. \u8bbe\u7f6eMapper\u7684\u7c7b\n        job.setMapperClass(WCMapper.class);\n\n        \/\/4. \u8bbe\u7f6eReducer\u7684\u7c7b\n        job.setReducerClass(WCReducer.class);\n\n        \/\/5. \u8bbe\u7f6eMapper\u8f93\u51fa\u7684\u7c7b\u578b\n        job.setMapOutputKeyClass(Text.class);\n        job.setMapOutputValueClass(LongWritable.class);\n\n        \/\/6. \u8bbe\u7f6eReducer\u8f93\u51fa\u7684\u7c7b\u578b\n        job.setOutputKeyClass(Text.class);\n        job.setOutputValueClass(LongWritable.class);\n\n\n        \/\/7. \u8bbe\u7f6ejob\u7684\u8f93\u5165\u8def\u5f84\n        FileInputFormat.setInputPaths(job, new Path(args[1]));\n        \/\/8. \u8bbe\u7f6ejob\u7684\u8f93\u51fa\u8def\u5f84\n        FileOutputFormat.setOutputPath(job, new Path(args[1]));\n\n\n        System.exit(job.waitForCompletion(true) ? 0 : 1);\n    }\n\n    \/\/                      mr\u8bfb\u53d6\u6587\u672c\u6587\u4ef6\u65f6 keyIn\u662fLongWritable\u7c7b\u578b\n    \/\/                              \u5f53\u524d\u884c\u5728\u6587\u4ef6\u4e2d\u7684\u5b57\u8282\u504f\u79fb\u91cf offset\n    \/\/                                    valueIn\u662fText\u7c7b\u578b\n    \/\/                               \u6587\u672c\u4e2d\u7684\u6bcf\u4e00\u884c\u5b57\u7b26\u4e32\n    \/\/               keyOut Text \u5b58\u50a8\u6bcf\u4e2a\u5355\u8bcd\n    \/\/               ValueOut LongWritable \u5b58\u50a8\u5355\u8bcd\u51fa\u73b0\u7684\u6570\u91cf\n    public static class WCMapper extends Mapper&lt;LongWritable, Text, Text, LongWritable&gt; {\n        private Text k = new Text();\n        private LongWritable v = new LongWritable(1);\n\n        @Override\n        protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {\n\n            \/\/ 1. \u53d6\u51fa\u6bcf\u884c\u7684\u5b57\u7b26\u4e32\n            String line = value.toString();\n            \/\/ 2. \u5207\u5206\u6210\u5355\u8bcd\n            String[] words = line.split(\"\\\\s+\");\n            \/\/ 3. \u7528context\u5c06\u5904\u7406\u5b8c\u7684\u7ed3\u679c\u5199\u51fa\u5230\u6846\u67b6\n            for (String word : words) {\n                word = word.replaceAll(\"\\\\W\", \"\").toLowerCase();\n                if (!\"\".equals(word)) {\n                    k.set(word);\n                    context.write(k, v);\n                }\n            }\n        }\n    }\n\n    public static class WCReducer extends Reducer&lt;Text, LongWritable, Text, LongWritable&gt; {\n        @Override\n        \/\/   MR\u6846\u67b6\u4f1a\u5c06Mapper\u8f93\u51fa\u7684\u6570\u636e\u6309\u7167Key\u8fdb\u884c\u5206\u7ec4\n        \/\/   \u5728\u8c03\u7528Reducer\u7684reduce\u65b9\u6cd5\u65f6\n        \/\/   \u6bcf\u6b21\u4f20\u5165\u4e00\u4e2akey \u548c\u8fd9\u4e2akey\u5bf9\u5e94\u7684\u6240\u6709Value\n        protected void reduce(Text key, Iterable&lt;LongWritable&gt; values, Context context) throws IOException, InterruptedException {\n            long count = 0L;\n            for (LongWritable value : values) {\n                count++;\n            }\n            context.write(key, new LongWritable(count));\n        }\n    }\n}\n<\/code><\/pre>\n<p>\u8fd0\u884c\u67b6\u5305<\/p>\n<pre><code class=\"language-bash line-numbers\">[root@wq1 bin]# pwd\n\/opt\/hadoop-2.7.7\/bin\n                        \u8981\u8fd0\u884c\u7684jar\u5305 \u8fd0\u884c\u7684\u4e3b\u7c7b  \u8981\u505a\u5904\u7406\u7684\u6587\u4ef6\u4f4d\u7f6e  \u5904\u7406\u540e\u8f93\u51fa\u7684\u4f4d\u7f6e\n[root@wq1 bin]# hadoop jar wc.jar MRWordCount \/wordcount\/input \/wordcount\/output\n<\/code><\/pre>\n<h1><span class=\"ez-toc-section\" id=\"%E4%B8%89%E3%80%81MapReduce%E7%9A%84%E6%A0%B8%E5%BF%83%E9%97%AE%E9%A2%98\"><\/span>\u4e09\u3001MapReduce\u7684\u6838\u5fc3\u95ee\u9898<span class=\"ez-toc-section-end\"><\/span><\/h1>\n<p>\u9996\u5148\u6211\u4eec\u8981\u660e\u786e\u4ed6\u7684\u5de5\u4f5c\u6d41\u7a0b\u662f\u4ec0\u4e48\u6837\u7684\uff0c\u901a\u5e38\u6211\u4eec\u8981\u9762\u4e34\u54ea\u4e9b\u95ee\u9898\uff0c\u901a\u8fc7\u53c2\u6570\u8c03\u4f18\u4f7f\u5f97\u673a\u5668\u80fd\u591f\u8fbe\u5230\u5f88\u597d\u7684\u72b6\u6001<\/p>\n<blockquote><p>\n  MapReduce\u5de5\u4f5c\u6d41\u7a0b\uff08\u5b98\u7f51\u7ed9\u51fa\u7684\u56fe\u7247\uff09\n<\/p><\/blockquote>\n<p><img src=\"https:\/\/img-blog.csdnimg.cn\/20201104214531460.png?\" alt=\"mapreduce\u5b98\u65b9\u5de5\u4f5c\u6d41\u7a0b\u56fe\" title=\"mapreduce\u5b98\u65b9\u5de5\u4f5c\u6d41\u7a0b\u56fe\" \/><\/p>\n<p>\u5408\u5e76\uff08Combine\uff09\u548c\u5f52\u5e76\uff08Merge\uff09\u7684\u533a\u522b\uff1a<br \/>\n\u4e24\u4e2a\u952e\u503c\u5bf9&lt;\u201ca\u201d,1>\u548c&lt;\u201ca\u201d,1>\uff0c\u5982\u679c\u5408\u5e76\uff0c\u4f1a\u5f97\u5230&lt;\u201ca\u201d,2>\uff0c\u5982\u679c\u5f52\u5e76\uff0c\u4f1a\u5f97\u5230&lt;\u201ca\u201d,&lt;1,1>><\/p>\n<p><span class=\"text-highlighted-inline\" style=\"background-color: #fffd38;\">\u5207\u7247 -> \u5206\u533apartition->\u5199\u5165\u73af\u5f62\u5185\u5b58\u7f13\u51b2\u533a->\u6267\u884c\u6ea2\u51fa\u5199 \u6392\u5e8fsort&#8212;>\u5408\u5e76combiner&#8212;>\u751f\u6210\u6ea2\u51fa\u5199\u6587\u4ef6->\u5f52\u5e76merge\u2014\u2014>reduce\u2014\u2014>\u590d\u5236copy->\u5f52\u5e76merge->reduce<\/span><\/p>\n<ol>\n<li>InputFormat\u6e90\u7801\u89e3\u6790<\/li>\n<\/ol>\n<ul>\n<li>\u627e\u5230\u4f60\u6570\u636e\u5b58\u50a8\u7684\u76ee\u5f55\u3002<\/li>\n<li>\u5f00\u59cb\u904d\u5386\u5904\u7406\uff08\u89c4\u5212\u5207\u7247\uff09\u76ee\u5f55\u4e0b\u7684\u6bcf\u4e00\u4e2a\u6587\u4ef6<\/li>\n<li>\u904d\u5386\u7b2c\u4e00\u4e2a\u6587\u4ef6test.txt\n<ol>\n<li>\u83b7\u53d6\u6587\u4ef6\u5927\u5c0ffs.sizeOf(test.txt);<\/li>\n<li>\u8ba1\u7b97\u5207\u7247\u5927\u5c0fcomputeSliteSize(Math.max(minSize,Math.min(maxSize,blocksize)))=blocksize=128M<\/li>\n<li>\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u5207\u7247\u5927\u5c0f=blocksize<\/li>\n<\/ol>\n<p>\u5f00\u59cb\u5207\uff0c\u5f62\u6210\u7b2c1\u4e2a\u5207\u7247\uff1atest.txt\u20140:128M \u7b2c2\u4e2a\u5207\u7247test.txt\u2014128:256M \u7b2c3\u4e2a\u5207\u7247test.txt\u2014256M:300M\uff08\u6bcf\u6b21\u5207\u7247\u65f6\uff0c\u90fd\u8981\u5224\u65ad\u5207\u5b8c\u5269\u4e0b\u7684\u90e8\u5206\u662f\u5426\u5927\u4e8e\u5757\u76841.1\u500d\uff0c\u4e0d\u5927\u4e8e1.1\u500d\u5c31\u5212\u5206\u4e00\u5757\u5207\u7247\uff09<br \/>\n\u5c06\u5207\u7247\u4fe1\u606f\u5199\u5230\u4e00\u4e2a\u5207\u7247\u89c4\u5212\u6587\u4ef6\u4e2d<br \/>\n\u6574\u4e2a\u5207\u7247\u7684\u6838\u5fc3\u8fc7\u7a0b\u5728getSplit()\u65b9\u6cd5\u4e2d\u5b8c\u6210\u3002<br \/>\n\u6570\u636e\u5207\u7247\u53ea\u662f\u5728\u903b\u8f91\u4e0a\u5bf9\u8f93\u5165\u6570\u636e\u8fdb\u884c\u5206\u7247\uff0c\u5e76\u4e0d\u4f1a\u518d\u78c1\u76d8\u4e0a\u5c06\u5176\u5207\u5206\u6210\u5206\u7247\u8fdb\u884c\u5b58\u50a8\u3002InputSplit\u53ea\u8bb0\u5f55\u4e86\u5206\u7247\u7684\u5143\u6570\u636e\u4fe1\u606f\uff0c\u6bd4\u5982\u8d77\u59cb\u4f4d\u7f6e\u3001\u957f\u5ea6\u4ee5\u53ca\u6240\u5728\u7684\u8282\u70b9\u5217\u8868\u7b49\u3002<br \/>\n\u6ce8\u610f\uff1ablock\u662fHDFS\u4e0a\u7269\u7406\u4e0a\u5b58\u50a8\u7684\u5b58\u50a8\u7684\u6570\u636e\uff0c\u5207\u7247\u662f\u5bf9\u6570\u636e\u903b\u8f91\u4e0a\u7684\u5212\u5206\u3002<br \/>\n\u63d0\u4ea4\u5207\u7247\u89c4\u5212\u6587\u4ef6\u5230yarn\u4e0a\uff0cyarn\u4e0a\u7684MrAppMaster\u5c31\u53ef\u4ee5\u6839\u636e\u5207\u7247\u89c4\u5212\u6587\u4ef6\u8ba1\u7b97\u5f00\u542fmaptask\u4e2a\u6570\u3002<\/p>\n<\/li>\n<\/ul>\n<ol start=\"2\">\n<li>\u5728\u5411\u8be5\u7f13\u51b2\u533a\u5199\u5165\u7684\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u5206\u533a\uff08partition\uff09<\/li>\n<\/ol>\n<p>\u4e5f\u5c31\u662f\u5bf9\u4e8e\u6bcf\u4e2a\u952e\u503c\u5bf9\u6765\u8bf4\uff0c\u90fd\u589e\u52a0\u4e86\u4e00\u4e2apartition\u5c5e\u6027\u503c\uff0c\u7136\u540e\u8fde\u540c\u952e\u503c\u5bf9\u4e00\u8d77\u5e8f\u5217\u5316\u6210\u5b57\u8282\u6570\u7ec4\u5199\u5165\u5230\u7f13\u51b2\u533a\uff08\u7f13\u51b2\u533a\u91c7\u7528\u7684\u5c31\u662f\u5b57\u8282\u6570\u7ec4\uff0c\u9ed8\u8ba4\u5927\u5c0f\u4e3a100M\uff09\u3002\u5206\u533a\u64cd\u4f5c\uff0c\u8fd9\u6837\u5c31\u80fd\u628amap\u4efb\u52a1\u5904\u7406\u7684\u7ed3\u679c\u53d1\u9001\u7ed9\u6307\u5b9a\u7684reducer\u53bb\u6267\u884c\uff0c\u4ece\u800c\u8fbe\u5230\u8d1f\u8f7d\u5747\u8861\uff0c\u907f\u514d\u6570\u636e\u503e\u659c\u3002MapReduce\u63d0\u4f9b\u9ed8\u8ba4\u7684\u5206\u533a\u7c7b\uff08HashPartitioner\uff09<\/p>\n<ol start=\"3\">\n<li>\u5f53\u5199\u5165\u7684\u6570\u636e\u91cf\u8fbe\u5230\u9884\u5148\u8bbe\u7f6e\u7684\u9619\u503c\u540e\uff08mapreduce.map.io.sort.spill.percent<\/li>\n<\/ol>\n<p>\u9ed8\u8ba40.80\uff0c\u6216\u800580%\uff09\u4fbf\u4f1a\u542f\u52a8\u6ea2\u5199\u51fa\u7ebf\u7a0b\u5c06\u7f13\u51b2\u533a\u4e2d\u7684\u90a3\u90e8\u5206\u6570\u636e\u6ea2\u51fa\u5199\uff08spill\uff09\u5230\u78c1\u76d8\u7684\u4e34\u65f6\u6587\u4ef6\u4e2d\uff0c\u5e76\u5728\u5199\u5165\u524d\u6839\u636ekey\u8fdb\u884c\u6392\u5e8f\uff08sort\uff09\u548c\u5408\u5e76\uff08combine\uff0c\u53ef\u9009\u64cd\u4f5c\uff09\u3002<\/p>\n<ol start=\"4\">\n<li>\u6ea2\u51fa\u5199\u8fc7\u7a0b\u6309\u8f6e\u8be2\u65b9\u5f0f\u5c06\u7f13\u51b2\u533a\u4e2d\u7684\u5185\u5bb9\u5199\u5230mapreduce.cluster.local.dir\u5c5e\u6027\u6307\u5b9a\u7684\u76ee\u5f55\u4e2d\u3002<\/p>\n<\/li>\n<li>\n<p>\u5f53\u6574\u4e2amap\u4efb\u52a1\u5b8c\u6210\u6ea2\u51fa\u5199\u540e<\/p>\n<\/li>\n<\/ol>\n<p>\u4f1a\u5bf9\u78c1\u76d8\u4e2d\u8fd9\u4e2amap\u4efb\u52a1\u4ea7\u751f\u7684\u6240\u6709\u4e34\u65f6\u6587\u4ef6\uff08spill\u6587\u4ef6\uff09\u8fdb\u884c\u5f52\u5e76\uff08merge\uff09\u64cd\u4f5c\u751f\u6210\u6700\u7ec8\u7684\u6b63\u5f0f\u8f93\u51fa\u6587\u4ef6\uff0c\u6b64\u65f6\u7684\u5f52\u5e76\u662f\u5c06\u6240\u6709spill\u6587\u4ef6\u4e2d\u7684\u76f8\u540cpartition\u5408\u5e76\u5230\u4e00\u8d77\uff0c\u5e76\u5bf9\u5404\u4e2apartition\u4e2d\u7684\u6570\u636e\u518d\u8fdb\u884c\u4e00\u6b21\u6392\u5e8f\uff08sort\uff09\uff0c\u751f\u6210key\u548c\u5bf9\u5e94\u7684value-list\uff0c\u6587\u4ef6\u5f52\u5e76\u65f6\uff0c\u5982\u679c\u6ea2\u5199\u6587\u4ef6\u6570\u91cf\u8d85\u8fc7\u53c2\u6570min.num.spills.for.combine\u7684\u503c\uff08\u9ed8\u8ba4\u4e3a3\uff09\u65f6\uff0c\u53ef\u4ee5\u518d\u6b21\u8fdb\u884c\u5408\u5e76\u3002\u81f3\u6b64\uff0cmap\u7aefshuffle\u8fc7\u7a0b\u7ed3\u675f\u3002<\/p>\n<ol start=\"6\">\n<li>Reduce\u8fdb\u7a0b\u542f\u52a8\u4e00\u4e9b\u6570\u636ecopy\u7ebf\u7a0b<\/li>\n<\/ol>\n<p>\u901a\u8fc7HTTP\u65b9\u5f0f\u8bf7\u6c42MapTask\u6240\u5728\u7684NodeManager\u4ee5\u83b7\u53d6\u8f93\u51fa\u6587\u4ef6\u3002NodeManager\u9700\u8981\u4e3a\u5206\u533a\u6587\u4ef6\u8fd0\u884creduce\u4efb\u52a1\u3002\u5e76\u4e14reduce\u4efb\u52a1\u9700\u8981\u96c6\u7fa4\u4e0a\u82e5\u5e72\u4e2amap\u4efb\u52a1\u7684map\u8f93\u51fa\u4f5c\u4e3a\u5176\u7279\u6b8a\u7684\u5206\u533a\u6587\u4ef6\u3002\u800c\u6bcf\u4e2amap\u4efb\u52a1\u7684\u5b8c\u6210\u65f6\u95f4\u53ef\u80fd\u4e0d\u540c\uff0c\u56e0\u6b64\u53ea\u8981\u6709\u4e00\u4e2a\u4efb\u52a1\u5b8c\u6210\uff0creduce\u4efb\u52a1\u5c31\u5f00\u59cb\u590d\u5236\u5176\u8f93\u51fa\u3002reduce\u4efb\u52a1\u6709\u5c11\u91cf\u590d\u5236\u7ebf\u7a0b\uff0c\u56e0\u6b64\u80fd\u591f\u5e76\u884c\u53d6\u5f97map\u8f93\u51fa\u3002\u9ed8\u8ba4\u7ebf\u7a0b\u6570\u4e3a5\uff0c\u4f46\u8fd9\u4e2a\u9ed8\u8ba4\u503c\u53ef\u4ee5\u901a\u8fc7mapreduce.reduce.shuffle.parallelcopies\u5c5e\u6027\u8fdb\u884c\u8bbe\u7f6e\u3002<\/p>\n<ol start=\"7\">\n<li>Copy\u8fc7\u6765\u7684\u6570\u636e\u4f1a\u5148\u653e\u5165\u5185\u5b58\u7f13\u51b2\u533a\u4e2d<\/li>\n<\/ol>\n<p>\u5982\u679c\u5185\u5b58\u7f13\u51b2\u533a\u4e2d\u80fd\u653e\u5f97\u4e0b\u8fd9\u6b21\u6570\u636e\u7684\u8bdd\u5c31\u76f4\u63a5\u628a\u6570\u636e\u5199\u5230\u5185\u5b58\u4e2d\uff0c\u5373\u5185\u5b58\u5230\u5185\u5b58merge\u3002Reduce\u8981\u5411\u6bcf\u4e2aMap\u53bb\u62d6\u53d6\u6570\u636e\uff0c\u5728\u5185\u5b58\u4e2d\u6bcf\u4e2aMap\u5bf9\u5e94\u4e00\u5757\u6570\u636e\uff0c\u5f53\u5185\u5b58\u7f13\u5b58\u533a\u4e2d\u5b58\u50a8\u7684Map\u6570\u636e\u5360\u7528\u7a7a\u95f4\u8fbe\u5230\u4e00\u5b9a\u7a0b\u5ea6\u7684\u65f6\u5019\uff0c\u5f00\u59cb\u542f\u52a8\u5185\u5b58\u4e2dmerge\uff0c\u628a\u5185\u5b58\u4e2d\u7684\u6570\u636emerge\u8f93\u51fa\u5230\u78c1\u76d8\u4e0a\u4e00\u4e2a\u6587\u4ef6\u4e2d\uff0c\u5373\u5185\u5b58\u5230\u78c1\u76d8merge\u3002\u4e0emap\u7aef\u7684\u6ea2\u5199\u7c7b\u4f3c\uff0c\u5728\u5c06buffer\u4e2d\u591a\u4e2amap\u8f93\u51fa\u5408\u5e76\u5199\u5165\u78c1\u76d8\u4e4b\u524d\uff0c\u5982\u679c\u8bbe\u7f6e\u4e86Combiner\uff0c\u5219\u4f1a\u5316\u7b80\u538b\u7f29\u5408\u5e76\u7684map\u8f93\u51fa\u3002Reduce\u7684\u5185\u5b58\u7f13\u51b2\u533a\u53ef\u901a\u8fc7mapred.job.shuffle.input.buffer.percent\u914d\u7f6e\uff0c\u9ed8\u8ba4\u662fJVM\u7684heap size\u768470%\u3002\u5185\u5b58\u5230\u78c1\u76d8merge\u7684\u542f\u52a8\u95e8\u9650\u53ef\u4ee5\u901a\u8fc7mapred.job.shuffle.merge.percent\u914d\u7f6e\uff0c\u9ed8\u8ba4\u662f66%\u3002<\/p>\n<ol start=\"8\">\n<li>\u5728reduce\u9636\u6bb5<\/li>\n<\/ol>\n<p>reduce()\u65b9\u6cd5\u7684\u8f93\u5165\u662f\u6240\u6709\u7684Key\u548c\u5b83\u7684Value\u8fed\u4ee3\u5668\u3002\u6b64\u9636\u6bb5\u7684\u8f93\u51fa\u76f4\u63a5\u5199\u5230\u8f93\u51fa\u6587\u4ef6\u7cfb\u7edf\uff0c\u4e00\u822c\u4e3aHDFS\u3002\u5982\u679c\u91c7\u7528HDFS\uff0c\u7531\u4e8eNodeManager\u4e5f\u8fd0\u884c\u6570\u636e\u8282\u70b9\uff0c\u6240\u4ee5\u7b2c\u4e00\u4e2a\u5757\u526f\u672c\u5c06\u88ab\u5199\u5230\u672c\u5730\u78c1\u76d8\u30021\u3001\u5f53reduce\u5c06\u6240\u6709\u7684map\u4e0a\u5bf9\u5e94\u81ea\u5df1partition\u7684\u6570\u636e\u4e0b\u8f7d\u5b8c\u6210\u540e\uff0creducetask\u771f\u6b63\u8fdb\u5165reduce\u51fd\u6570\u7684\u8ba1\u7b97\u9636\u6bb5\u3002\u7531\u4e8ereduce\u8ba1\u7b97\u65f6\u540c\u6837\u662f\u9700\u8981\u5185\u5b58\u4f5c\u4e3abuffer\uff0c\u53ef\u4ee5\u7528mapreduce.reduce.input.buffer.percent\uff08default 0.0\uff09(\u6e90\u4ee3\u7801MergeManagerImpl.java\uff1a674\u884c)\u6765\u8bbe\u7f6ereduce\u7684\u7f13\u5b58\u3002\u8fd9\u4e2a\u53c2\u6570\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u4e3a0\uff0c\u4e5f\u5c31\u662f\u8bf4\uff0creduce\u662f\u5168\u90e8\u4ece\u78c1\u76d8\u5f00\u59cb\u8bfb\u5904\u7406\u6570\u636e\u3002\u5982\u679c\u8fd9\u4e2a\u53c2\u6570\u5927\u4e8e0\uff0c\u90a3\u4e48\u5c31\u4f1a\u6709\u4e00\u5b9a\u91cf\u7684\u6570\u636e\u88ab\u7f13\u5b58\u5728\u5185\u5b58\u5e76\u8f93\u9001\u7ed9reduce\uff0c\u5f53reduce\u8ba1\u7b97\u903b\u8f91\u6d88\u8017\u5185\u5b58\u5f88\u5c0f\u65f6\uff0c\u53ef\u4ee5\u5206\u4e00\u90e8\u5206\u5185\u5b58\u7528\u6765\u7f13\u5b58\u6570\u636e\uff0c\u53ef\u4ee5\u63d0\u5347\u8ba1\u7b97\u7684\u901f\u5ea6\u3002\u6240\u4ee5\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u90fd\u662f\u4ece\u78c1\u76d8\u8bfb\u53d6\u6570\u636e\uff0c\u5982\u679c\u5185\u5b58\u8db3\u591f\u5927\u7684\u8bdd\uff0c\u52a1\u5fc5\u8bbe\u7f6e\u8be5\u53c2\u6570\u8ba9reduce\u76f4\u63a5\u4ece\u7f13\u5b58\u8bfb\u6570\u636e\uff0c\u8fd9\u6837\u505a\u5c31\u6709\u70b9Spark Cache\u7684\u611f\u89c9\u3002<\/p>\n<blockquote><p>\n  map task\u6570\u91cf\uff0cpartition\u6570\u91cf\uff0creduce task\u6570\u91cf\u600e\u4e48\u786e\u5b9a\n<\/p><\/blockquote>\n<p>\u6709\u65f6\u5019\u9700\u8981\u628a\u6700\u7ec8\u7684\u8f93\u51fa\u6570\u636e\u5206\u5230\u4e0d\u540c\u7684\u6587\u4ef6\u4e2d\uff0c\u6bd4\u5982\u6309\u7167\u7701\u4efd\u5212\u5206\u7684\u8bdd\uff0c\u9700\u8981\u628a\u540c\u4e00\u7701\u4efd\u7684\u6570\u636e\u653e\u5230\u4e00\u4e2a\u6587\u4ef6\u4e2d.\u6211\u4eec\u77e5\u9053\u6700\u7ec8\u7684\u8f93\u51fa\u6570\u636e\u662f\u6765\u81ea\u4e8eReducer\u4efb\u52a1\u3002\u90a3\u4e48\uff0c\u5982\u679c\u8981\u5f97\u5230\u591a\u4e2a\u6587\u4ef6\uff0c\u610f\u5473\u7740\u6709\u540c\u6837\u6570\u91cf\u7684Reducer\u4efb\u52a1\u5728\u8fd0\u884c\u3002Reducer\u4efb\u52a1\u7684\u6570\u636e\u6765\u81ea\u4e8eMapper\u4efb\u52a1\uff0c\u4e5f\u5c31\u8bf4Mapper\u4efb\u52a1\u8981\u5212\u5206\u6570\u636e\uff0c\u5bf9\u4e8e\u4e0d\u540c\u7684\u6570\u636e\u5206\u914d\u7ed9\u4e0d\u540c\u7684Reducer\u4efb\u52a1\u8fd0\u884c\u3002Mapper\u4efb\u52a1\u5212\u5206\u6570\u636e\u7684\u8fc7\u7a0b\u5c31\u79f0\u4f5cPartition\u3002\u8d1f\u8d23\u5b9e\u73b0\u5212\u5206\u6570\u636e\u7684\u7c7b\u79f0\u4f5cPartitioner\u3002<\/p>\n<ol>\n<li>map \u6570\u91cf<\/li>\n<\/ol>\n<p>\u4e00\u4e2atask\u7684map\u6570\u91cf\u7531splitSize\u6765\u51b3\u5b9a\u7684,\u4e5f\u5c31\u662f\u7531\u5757\u5927\u5c0f\u6765\u51b3\u5b9a<\/p>\n<pre><code class=\"language-java line-numbers\">goalSize = totalSize \/ mapred.map.tasks\ninSize = max {mapred.min.split.size, minSplitSize}\nsplitSize = max (minSize, min(goalSize, dfs.block.size))\n<\/code><\/pre>\n<ol start=\"2\">\n<li>\u5206\u533a\u6570\u91cf<\/li>\n<\/ol>\n<ul>\n<li>\u4e4b\u524d\u7684\u65e5\u5fd7\u5206\u6790\u9879\u76ee\uff0c\u8981\u6c42\u5c06\u6bcf\u5929\u7684\u65e5\u5fd7\u5904\u7406\u5355\u72ec\u8f93\u51fa\u5230\u4e00\u4e2a\u6587\u4ef6\u4e2d\uff0c\u53ef\u4ee5\u5229\u7528\u5206\u533a\u6765\u5b9e\u73b0<\/li>\n<li>\u5047\u8bbe\u81ea\u5b9a\u4e49\u5206\u533a\u6570\u4e3a5\uff0c\u5219<br \/>\njob.setNumReduceTasks(1);\u4f1a\u6b63\u5e38\u8fd0\u884c\uff0c\u53ea\u4e0d\u8fc7\u4f1a\u4ea7\u751f\u4e00\u4e2a\u8f93\u51fa\u6587\u4ef6<br \/>\njob.setNumReduceTasks(2);\u4f1a\u62a5\u9519<br \/>\njob.setNumReduceTasks(6);\u5927\u4e8e5\uff0c\u7a0b\u5e8f\u4f1a\u6b63\u5e38\u8fd0\u884c\uff0c\u4f1a\u4ea7\u751f\u7a7a\u6587\u4ef6<\/li>\n<\/ul>\n<p>\u6e90\u7801\u5206\u6790\uff0c\u53ef\u901a\u8fc7\u81ea\u5b9a\u4e49\u5206\u533a\u65b9\u6cd5\u5b9e\u73b0\u5206\u533a<\/p>\n<pre><code class=\"language-java line-numbers\">package org.apache.hadoop.mapreduce.lib.partition;\n\nimport org.apache.hadoop.mapreduce.Partitioner;\n\n\/** Partition keys by their {@link Object#hashCode()}. *\/\npublic class HashPartitioner&lt;K, V&gt; extends Partitioner&lt;K, V&gt; {\n\n  \/** Use {@link Object#hashCode()} to partition. *\/\n  public int getPartition(K key, V value,int numReduceTasks) {\n    \/\/\u9ed8\u8ba4\u4f7f\u7528key\u7684hash\u503c\u4e0e\u4e0aint\u7684\u6700\u5927\u503c\uff0c\u907f\u514d\u51fa\u73b0\u6570\u636e\u6ea2 \u7684\u60c5\u51b5\n    return (key.hashCode() &amp; Integer.MAX_VALUE) % numReduceTasks;\n  }\n\n}\n\n<\/code><\/pre>\n<ol start=\"3\">\n<li>reduce\u6570\u91cf<br \/>\n\u4e00\u4e2atask\u7684reduce\u6570\u91cf\uff0c\u7531partition\u51b3\u5b9a<\/li>\n<\/ol>\n<pre><code class=\"language-java line-numbers\">        job.setPartitionerClass(PartitionChoice.class);\n        job.setNumReduceTasks(3);\n<\/code><\/pre>\n<hr \/>\n<blockquote><p>\n  \u4f7f\u7528mapreduce\u8fdb\u884c\u5355\u8bcd\u8ba1\u6570\u7684\u8fc7\u7a0b\n<\/p><\/blockquote>\n<p><img src=\"https:\/\/img-blog.csdnimg.cn\/20201105002604553.png\" alt=\"mapreduce\u7b80\u56fe\" title=\"mapreduce\u7b80\u56fe\" \/><\/p>\n<hr \/>\n<h3><span class=\"ez-toc-section\" id=\"MapReduce%E7%9A%84%E5%87%A0%E7%A7%8D%E6%8E%92%E5%BA%8F%E6%96%B9%E5%BC%8F\"><\/span>MapReduce\u7684\u51e0\u79cd\u6392\u5e8f\u65b9\u5f0f<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<ol>\n<li>\u90e8\u5206\u6392\u5e8f\uff1a\n<p>MapReduce\u6839\u636e\u8f93\u5165\u8bb0\u5f55\u7684\u952e\u5bf9\u6570\u636e\u96c6\u6392\u5e8f\u3002\u4fdd\u8bc1\u8f93\u51fa\u7684\u6bcf\u4e2a\u6587\u4ef6\u5185\u90e8\u6392\u5e8f\u3002<\/p>\n<\/li>\n<li>\n<p>\u5168\u6392\u5e8f\uff1a<\/p>\n<p>\u5982\u4f55\u7528Hadoop\u4ea7\u751f\u4e00\u4e2a\u5168\u5c40\u6392\u5e8f\u7684\u6587\u4ef6\uff1f\u6700\u7b80\u5355\u7684\u65b9\u6cd5\u662f\u4f7f\u7528\u4e00\u4e2a\u5206\u533a\u3002\u4f46\u8be5\u65b9\u6cd5\u5728\u5904\u7406\u5927\u578b    \u6587\u4ef6\u65f6\u6548\u7387\u6781\u4f4e\uff0c\u56e0\u4e3a\u4e00\u53f0\u673a\u5668\u5fc5\u987b\u5904\u7406\u6240\u6709\u8f93\u51fa\u6587\u4ef6\uff0c\u4ece\u800c\u5b8c\u5168\u4e27\u5931\u4e86MapReduce\u6240\u63d0\u4f9b\u7684   \u5e76\u884c\u67b6\u6784\u3002<br \/>\n\u66ff\u4ee3\u65b9\u6848\uff1a\u9996\u5148\u521b\u5efa\u4e00\u7cfb\u5217\u6392\u597d\u5e8f\u7684\u6587\u4ef6\uff1b\u5176\u6b21\uff0c\u4e32\u8054\u8fd9\u4e9b\u6587\u4ef6\uff1b\u6700\u540e\uff0c\u751f\u6210\u4e00\u4e2a\u5168\u5c40\u6392\u5e8f\u7684   \u6587\u4ef6\u3002\u4e3b\u8981\u601d\u8def\u662f\u4f7f\u7528\u4e00\u4e2a\u5206\u533a\u6765\u63cf\u8ff0\u8f93\u51fa\u7684\u5168\u5c40\u6392\u5e8f\u3002<\/p>\n<\/li>\n<li>\n<p>\u8f85\u52a9\u6392\u5e8f\uff1a\uff08GroupingComparator\u5206\u7ec4\uff09<\/p>\n<p>Mapreduce\u6846\u67b6\u5728\u8bb0\u5f55\u5230\u8fbereducer\u4e4b\u524d\u6309\u952e\u5bf9\u8bb0\u5f55\u6392\u5e8f\uff0c\u4f46\u952e\u6240\u5bf9\u5e94\u7684\u503c\u5e76\u6ca1\u6709\u88ab\u6392\u5e8f\u3002\u751a\u81f3   \u5728\u4e0d\u540c\u7684\u6267\u884c\u8f6e\u6b21\u4e2d\uff0c\u8fd9\u4e9b\u503c\u7684\u6392\u5e8f\u4e5f\u4e0d\u56fa\u5b9a\uff0c\u56e0\u4e3a\u5b83\u4eec\u6765\u81ea\u4e0d\u540c\u7684map\u4efb\u52a1\u4e14\u8fd9\u4e9bmap\u4efb\u52a1 \u5728\u4e0d\u540c\u8f6e\u6b21\u4e2d\u5b8c\u6210\u65f6\u95f4\u5404\u4e0d\u76f8\u540c\u3002\u4e00\u822c\u6765\u8bf4\uff0c\u5927\u591a\u6570MapReduce\u7a0b\u5e8f\u4f1a\u907f\u514d\u8ba9reduce\u51fd\u6570\u4f9d\u8d56    \u4e8e\u503c\u7684\u6392\u5e8f\u3002\u4f46\u662f\uff0c\u6709\u65f6\u4e5f\u9700\u8981\u901a\u8fc7\u7279\u5b9a\u7684\u65b9\u6cd5\u5bf9\u952e\u8fdb\u884c\u6392\u5e8f\u548c\u5206\u7ec4\u7b49\u4ee5\u5b9e\u73b0\u5bf9\u503c\u7684\u6392\u5e8f\u3002<\/p>\n<\/li>\n<li>\n<p>\u4e8c\u6b21\u6392\u5e8f\uff1a<\/p>\n<p>\u5728\u81ea\u5b9a\u4e49\u6392\u5e8f\u8fc7\u7a0b\u4e2d\uff0c\u5982\u679ccompareTo\u4e2d\u7684\u5224\u65ad\u6761\u4ef6\u4e3a\u4e24\u4e2a\u5373\u4e3a\u4e8c\u6b21\u6392\u5e8f\u3002<\/p>\n<\/li>\n<\/ol>\n<hr \/>\n<blockquote>\n<p>Shuffle \u4ea7\u751f\u7684\u610f\u4e49\u662f\u4ec0\u4e48\uff1f\n<\/p><\/blockquote>\n<p>\u5b8c\u6574\u5730\u4ece map task \u7aef\u62c9\u53d6\u6570\u636e\u5230 reduce \u7aef\uff1b\u5728\u8de8\u8282\u70b9\u62c9\u53d6\u6570\u636e\u65f6\uff0c\u5c3d\u53ef\u80fd\u5730\u51cf\u5c11\u5bf9\u5e26\u5bbd\u7684\u4e0d\u5fc5\u8981\u6d88\u8017\uff1b\u51cf\u5c11\u78c1\u76d8 IO \u5bf9 task \u6267\u884c\u7684\u5f71\u54cd\uff1b<br \/>\n\u6bcf\u4e2a map task \u90fd\u6709\u4e00\u4e2a\u5185\u5b58\u7f13\u51b2\u533a\uff0c\u5b58\u50a8\u7740 map \u7684\u8f93\u51fa\u7ed3\u679c<br \/>\n\u5f53\u7f13\u51b2\u533a\u5feb\u6ee1\u7684\u65f6\u5019\u9700\u8981\u5c06\u7f13\u51b2\u533a\u7684\u6570\u636e\u4ee5\u4e00\u4e2a\u4e34\u65f6\u6587\u4ef6\u7684\u65b9\u5f0f\u5b58\u653e\u5230\u78c1\u76d8\uff0c\u5f53\u6574\u4e2amap task \u7ed3\u675f\u540e\u518d\u5bf9\u78c1\u76d8\u4e2d\u8fd9\u4e2a map task \u4ea7\u751f\u7684\u6240\u6709\u4e34\u65f6\u6587\u4ef6\u505a\u5408\u5e76\uff0c\u751f\u6210\u6700\u7ec8\u7684\u6b63\u5f0f\u8f93\u51fa\u6587\u4ef6\uff0c\u7136\u540e\u7b49\u5f85 reduce task \u6765\u62c9\u6570\u636e\u3002<\/p>\n<ul>\n<li>\u7279\u6b8a\u60c5\u51b5\uff1a\u5f53\u6570\u636e\u91cf\u5f88\u5c0f\uff0c\u8fbe\u4e0d\u5230\u7f13\u51b2\u533a\u9619\u503c\u65f6\uff0c\u600e\u4e48\u5904\u7406\uff1f<br \/>\n\u4e0d\u4f1a\u6709\u5199\u4e34\u65f6\u6587\u4ef6\u5230\u78c1\u76d8\u7684\u64cd\u4f5c\uff0c\u4e5f\u4e0d\u4f1a\u6709\u540e\u9762\u7684\u5408\u5e76,\u65e0reduce\u4efb\u52a1\u3002<\/li>\n<\/ul>\n<hr \/>\n<blockquote><p>\n  MapReduce \u600e\u4e48\u5b9e\u73b0 TopN\uff1f\n<\/p><\/blockquote>\n<p>\u5b9e\u73b0WritableComparable\u63a5\u53e3<\/p>\n<pre><code class=\"language-java line-numbers\">public class compares implements WritableComparable&lt;compares&gt; {\n@Override\n    public int compareTo(FlowBean o) {\n        if (this.sumFlow &gt; o.getSumFlow()){\n            return -1;\n        }else if (this.sumFlow &lt; o.getSumFlow()){\n            return 1;\n        }else {\n            return 0;\n        }\n    }\n}   \n\/\/\u90a3\u4e48\u5728map\u548creduce\u4e2d\u518d\u901a\u8fc7\u4e0b\u9762\u8fd9\u65b9\u6cd5\u5b9e\u73b0topn\ncompares = new compares();\n if (kBeans.size() &gt; n){\n            kBeans.remove(kBeans.lastKey());\n        }\n<\/code><\/pre>\n<h3><span class=\"ez-toc-section\" id=\"MapReduce%E8%B0%83%E4%BC%98\"><\/span>MapReduce\u8c03\u4f18<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<ol>\n<li>\u5173\u4e8e\u786c\u4ef6\u65b9\u9762\u7684\u8c03\u4f18<br \/>\n\u5185\u5b58+\u786c\u76d8<br \/>\n\u786c\u4ef6\u7684\u4e3b\u8981\u914d\u7f6e\u6709 \uff1amemory   \u5f53CPU\u4e0d\u591f\u7684\u8bdd \uff1a \u7a0b\u5e8f\u6267\u884c\u8fd0\u7b97\u4f1a\u6162\u4e00\u70b9\uff0c\u5f53\u5185\u5b58\u4e0d\u591f\u7684 \uff1a \u4f1a\u5bfc\u81f4heap\u6ea2\u51fa\uff0cGC\u3002\u56e0\u4e3a\u5728\u5185\u5b58\u4e0d\u591f\u7684\u65f6\u5019\uff0cYarn\u91cc\u9762\u6709\u4e00\u4e2a\u52a8\u6001\u7684\u7ebf\u7a0b\u5c31\u4f1a\u68c0\u6d4b\u5230 \uff0c \u7136\u540e\u5c31\u7ed9\u4f60\u6740\u6b7b\u4e86<\/li>\n<li>Linux\u5c42\u9762\u7684\u4e00\u4e9b\u8c03\u4f18<br \/>\nvim \/etc\/security\/limits.conf<\/li>\n<\/ol>\n<pre><code class=\"language-bash line-numbers\">    soft nofile 65535     \u5355\u4e2a\u7528\u6237\u53ef\u7528\u7684\u6700\u5927\u8fdb\u7a0b\u6570\u91cf\uff08\u8f6f\u9650\u5236\uff09\n    hard nofile 65535    \u5355\u4e2a\u7528\u6237\u53ef\u7528\u7684\u6700\u5927\u8fdb\u7a0b\u6570\u91cf\uff08\u786c\u9650\u5236\uff09\n\n    soft nproc 65535     \u53ef\u6253\u5f00\u7684\u6587\u4ef6\u63cf\u8ff0\u7b26\u7684\u6700\u5927\u6570\uff08\u8f6f\u9650\u5236\uff09\n    hard nproc 65535    \u53ef\u6253\u5f00\u7684\u6587\u4ef6\u63cf\u8ff0\u7b26\u7684\u6700\u5927\u6570\uff08\u786c\u9650\u5236\uff09\n<\/code><\/pre>\n<ol start=\"3\">\n<li>NameNode\u4e0aJVM\u7684\u53c2\u6570\u8c03\u4f18<\/li>\n<li>\u914d\u7f6e\u65b9\u9762\u7684\u8c03\u4f18<\/li>\n<\/ol>\n<h4><span class=\"ez-toc-section\" id=\"core-sitexml_%E7%9A%84%E8%B0%83%E6%95%B4\"><\/span>core-site.xml \u7684\u8c03\u6574<span class=\"ez-toc-section-end\"><\/span><\/h4>\n<ul>\n<li>Ipc.server.listen.queue.size \u63a7\u5236\u4e86\u670d\u52a1\u7aefsocket\u7684\u76d1\u542c\u961f\u5217\u957f\u5ea6<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Ipc.server.listen.queue.size&lt;\/name&gt;\n      &lt;value&gt;65535&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>o.file.buffer.size\u9ed8\u8ba4\u503c4096\uff084K\uff09\uff0c\u4f5c\u4e3ahadoop\u7684\u7f13\u51b2\u533a\uff0c\u7528\u4e8ehdfs\u6587\u4ef6\u7684\u8bfb\u5199\u3002\u8fd8\u6709map\u7684\u8f93\u51fa\u90fd\u7528\u5230\u8fd9\u4e2a\u7f13\u51b2\u533a\uff0c\u8f83\u5927\u7684\u7f13\u5b58\u80fd\u63d0\u9ad8\u6570\u636e\u4f20\u8f93\u8fd9\u662f\u8bfb\u5199sequence file\u7684buffer size\uff0c\u53ef\u51cf\u5c11I\/O\u6b21\u6570\u3002\u5982\u679c\u96c6\u7fa4\u6bd4\u8f83\u5927\uff0c\u5efa\u8bae\u628a\u53c2\u6570\u8c03\u523065535-131072<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Io.file.buffer.size&lt;\/name&gt;\n      &lt;value&gt;65535&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Hdfs\u7684\u56de\u6536\u7ad9\uff0c\u4e00\u4f46\u64cd\u4f5c\u4e8b\u7269\uff0c\u5220\u6389\u4e86HDFS\u4e2d\u7684\u6570\u636e\uff0c\u53ef\u4ee5\u627e\u56de\u3002\u9ed8\u8ba4\u4e0d\u5f00\u542f<br \/>\nValue\uff1a1440\uff08\u5206\u949f\uff091\u5929<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;fs.trash.interval&lt;\/name&gt;\n      &lt;value&gt;1440&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<hr \/>\n<h4><span class=\"ez-toc-section\" id=\"hdfs-sitexml_%E9%85%8D%E7%BD%AE%E6%96%87%E4%BB%B6%E7%9A%84%E8%B0%83%E6%95%B4\"><\/span>hdfs-site.xml \u914d\u7f6e\u6587\u4ef6\u7684\u8c03\u6574<span class=\"ez-toc-section-end\"><\/span><\/h4>\n<ul>\n<li>HDFS\u4e2d\u7684block\u5757\u5927\u5c0f\u7684\u8bbe\u7f6e<br \/>\n\u6709\u65f6\u5019\u6211\u4eec\u573a\u666f\u4e2d\u53ef\u80fd\u51fa\u73b0\u5927\u91cf\u5c0f\u6587\u4ef6\uff0c\u8fd9\u65f6\u5019\u6211\u4eec\u53ef\u4ee5\u9002\u5f53\u7684\u8c03\u5c0f\uff0c\u6bd4\u598216M(16777216)<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;dfs.blockSize&lt;\/name&gt;\n      &lt;value&gt;134217728&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u5e26\u5bbd:\u9ed8\u8ba4\u662f1M\uff08\u5728balancer\u7684\u65f6\u5019\uff0c\u8bbe\u7f6ehdfs\u4e2d\u6570\u636e\u79fb\u52a8\u901f\u5ea6\uff09\u96c6\u7fa4\u4e00\u822c\u662f\u5343M\u8def\u7531\u5668\uff0c\u6240\u4ee5\u5c3d\u91cf\u6539\u5927<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;dfs.datanode.balance.bandwidthPerSec&lt;\/name&gt;\n      &lt;value&gt;1048576&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u771f\u6b63datanode\u6570\u636e\u4fdd\u5b58\u8def\u5f84\uff0c\u53ef\u4ee5\u5199\u591a\u5757\u786c\u76d8\u3002\u4e3b\u8981\u7528\u6765\u5b9e\u73b0IO\u5e73\u8861\uff0c\u56e0\u6b64\u4f1a\u663e\u8457\u6539\u8fdb\u78c1\u76d8IO\u6027\u80fd<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;dfs.datanode.data.dir&lt;\/name&gt;\n      &lt;value&gt;\/sda,\/sda2,\/sda3&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Namenode server RPC \u7684\u5904\u7406\u7ebf\u7a0b\u6570\uff0c\u9ed8\u8ba4\u662f10\uff0cnamenode\u7ebf\u7a0b\u901a\u8fc7RPC\u7684\u65b9\u5f0f\u8ddfdatanode\u901a\u4fe1\uff0c\u5982\u679cdatanode\u6570\u91cf\u592a\u591a\u65f6\u53ef\u80fd\u51fa\u73b0RPC timeout,\u63d0\u5347\u7f51\u7edc\u901f\u5ea6\u6216\u8005\u63d0\u9ad8\u8fd9\u4e2a\u503c\u3002\u4f46\u662fthread\u6570\u91cf\u589e\u591a\u4e5f\u4ee3\u8868\u7740namenode\u6d88\u8017\u7684\u5185\u5b58\u4e5f\u968f\u7740\u589e\u52a0<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Dfs.namenode.hadler.count&lt;\/name&gt;\n      &lt;value&gt;30&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<hr \/>\n<h4><span class=\"ez-toc-section\" id=\"mapred-sitexml\"><\/span>mapred-site.xml<span class=\"ez-toc-section-end\"><\/span><\/h4>\n<ul>\n<li>3  \u9ed8\u8ba4\u5f00\u542f\u7684reduce\u6570\u91cf\uff0c\u591a\u53f0\u673a\u5668\u5206\u62c5\u4e00\u53f0\u673a\u5668\u7684\u538b\u529b<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Mapreduce.job.reduces&lt;\/name&gt;\n      &lt;value&gt;3&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u73af\u5f62\u7f13\u51b2\u533a\u6240\u5360\u5185\u5b58\u5927\u5c0f\u9ed8\u8ba4100M<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt; Mapreduce.task.io.sort.mb&lt;\/name&gt;\n      &lt;value&gt;200&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u73af\u5f62\u7f13\u51b2\u533a\u7684\u9608\u503c  \u9ed8\u8ba4\u7684\u662f0.8<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Mapreduce.map.sort.spill.percent&lt;\/name&gt;\n      &lt;value&gt;0.8&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Reduce Task\u4e2d\u5408\u5e76\u5c0f\u6587\u4ef6\u65f6\uff0c\u4e00\u6b21\u5408\u5e76\u7684\u6587\u4ef6\u6570\u636e\uff0c\u6bcf\u6b21\u5408\u5e76\u7684\u65f6\u5019\u9009\u62e9\u6700\u5c0f\u7684\u524d10\uff08\u9ed8\u8ba4\u503c\uff09\u8fdb\u884c\u5408\u5e76\u3002<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Mapreduce.task.io.sort.factor&lt;\/name&gt;\n      &lt;value&gt;50&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>map\u8f93\u51fa\u662f\u5426\u8fdb\u884c\u538b\u7f29\uff0c\u5982\u679c\u538b\u7f29\u5c31\u4f1a\u591a\u8017cpu\uff0c\u4f46\u662f\u51cf\u5c11\u4f20\u8f93\u65f6\u95f4\uff0c\u5982\u679c\u4e0d\u538b\u7f29\uff0c\u5c31\u9700\u8981\u8f83\u591a\u7684\u4f20\u8f93\u5e26\u5bbd\u3002\u9700\u8981\u914d   \u5408mapreduce.map.output.compress.codec\uff08\u6307\u5b9a\u538b\u7f29\u65b9\u5f0f\uff09<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Mapreduce.map.output.compress&lt;\/name&gt;\n      &lt;value&gt;true&lt;\/value&gt;\n&lt;\/property&gt;\n#\u538b\u7f29\u65b9\u5f0f\uff08\u727a\u7272CPU\u6362IO\u548c\u78c1\u76d8\u7684\u65b9\u5f0f\uff09\n  &lt;property&gt;\n      &lt;name&gt;Mapreduce.map.output.compress.codec&lt;\/name&gt;\n      &lt;value&gt;org.apache.hadoop.io.compress.SnappyCodec&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Reduce shuffle\u9636\u6bb5\u5e76\u884c\u4f20\u8f93\u7684\u6570\u91cf\u3002\u6839\u636e\u96c6\u7fa4\u5927\u5c0f\u53ef\u8c03<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;mapreduce.reduce.shuffle.parallelcopies&lt;\/name&gt;\n      &lt;value&gt;20&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Map\u548creduce\u662f\u901a\u8fc7http\u4f20\u8f93\u7684\uff0c\u8fd9\u4e2a\u662f\u8bbe\u7f6e\u4f20\u8f93\u7684\u5e76\u884c\u6570<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Mapreduce.tasktracker.http.threads&lt;\/name&gt;\n      &lt;value&gt;40&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u5bb9\u9519\u6027\u76f8\u5173\u7684\u53c2\u6570<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">#\u6bcf\u4e2aMap Task\u6700\u5927\u91cd\u8bd5\u6b21\u6570\uff0c\u4e00\u65e6\u91cd\u8bd5\u53c2\u6570\u8d85\u8fc7\u8be5\u503c\uff0c\u5219\u8ba4\u4e3a Map Task \u8fd0\u884c\u5931\u8d25\u9ed8\u8ba44\n  &lt;property&gt;\n      &lt;name&gt;mapreduce.map.maxattempts&lt;\/name&gt;\n      &lt;value&gt;4&lt;\/value&gt;\n&lt;\/property&gt;\n#\u6bcf\u4e2aReduce Task\u6700\u5927\u91cd\u8bd5\u6b21\u6570\uff0c\u4e00\u65e6\u91cd\u8bd5\u53c2\u6570\u8d85\u8fc7\u8be5\u503c\uff0c\u5219\u8ba4\u4e3a Map Task \u8fd0\u884c\u5931\u8d25\uff0c\u9ed8\u8ba44\n\n  &lt;property&gt;\n      &lt;name&gt;mapreduce.reduce.maxattempts&lt;\/name&gt;\n      &lt;value&gt;4&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<hr \/>\n<h4><span class=\"ez-toc-section\" id=\"Yarn-sitexml%E9%85%8D%E7%BD%AE\"><\/span>Yarn-site.xml\u914d\u7f6e<span class=\"ez-toc-section-end\"><\/span><\/h4>\n<ul>\n<li>\u7ed9nodemanager\u53ef\u7528\u7684\u7269\u7406\u5185\u5b58(8GB)<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;yarn.nodemanager.resource.memory-mb&lt;\/name&gt;\n      &lt;value&gt;8192&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u5355\u4e2a\u4efb\u52a1\u53ef\u7533\u8bf7\u7684\u6700\u5c11\u5185\u5b58 \u4e5f\u5c31\u662fRM \u4e2d\u6bcf\u4e2a\u5bb9\u5668\u8bf7\u6c42\u7684\u6700\u5c0f\u914d\u7f6e<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Yarn.scheduler.minimum-allocation-mb&lt;\/name&gt;\n      &lt;value&gt;1024&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>\u5355\u4e2a\u4efb\u52a1\u53ef\u7533\u8bf7\u7684\u6700\u5927\u5185\u5b58<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;yarn.scheduler.maximum-allocation-mb&lt;\/name&gt;\n      &lt;value&gt;8192&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<ul>\n<li>Yarn\u8fd9\u4e2a\u8282\u70b9\u53ef\u4f7f\u7528\u7684\u865a\u62dfCPU\u4e2a\u6570 \u9ed8\u8ba4\u662f8.\u4f46\u6211\u4eec\u901a\u5e38\u914d\u7f6e\u6210\u8ddf\u7269\u7406CPU\u4e2a\u6570\u4e00\u6837<\/li>\n<\/ul>\n<pre data-language=XML><code class=\"language-markup line-numbers\">  &lt;property&gt;\n      &lt;name&gt;Yarn.nodemanager.resource.cpu-vcores&lt;\/name&gt;\n      &lt;value&gt;4&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<hr \/>\n<h4><span class=\"ez-toc-section\" id=\"%E9%87%8D%E8%A6%81%E7%9A%84%E5%8F%82%E6%95%B0%E9%85%8D%E7%BD%AE\"><\/span>\u91cd\u8981\u7684\u53c2\u6570\u914d\u7f6e<span class=\"ez-toc-section-end\"><\/span><\/h4>\n<pre data-language=XML><code class=\"language-markup line-numbers\">   #\u9ed8\u8ba4\u4e3a 1024\u3002\u5982\u679c Map Task \u5b9e\u9645\u4f7f\u7528\u7684\u8d44\u6e90\u91cf\u8d85\u8fc7\u8be5\u503c\uff0c\u5219\u4f1a\u88ab\u5f3a\u5236\u6740\u6b7b\u3002\u53ef\u4ee5\u9002\u5f53\u7684\u8c03\n  &lt;property&gt;\n      &lt;name&gt;mapreduce.map.memory.mb&lt;\/name&gt;\n      &lt;value&gt;1280&lt;\/value&gt;\n&lt;\/property&gt;\n# Map\u7aef\u542f\u7528JVM\u6240\u5360\u7528\u7684\u5185\u5b58\u5927\u5c0f  \uff08\u5b98\u65b9-----&gt;200M  heap over\/GC\uff09\n# -XX:-UseGCOverheadLimit\uff08\u7981\u7528GC,\u6211\u4eec\u4f7f\u7528parNew+CMS\u65b9\u5f0f\uff09\n  &lt;property&gt;\n      &lt;name&gt;Mapreduce.map.java.opts&lt;\/name&gt;\n      &lt;value&gt;-Xmx1024m -XX:-UseGCOverheadLimit&lt;\/value&gt;\n&lt;\/property&gt;\n#  reduce\u9636\u6bb5\uff0creduce\u7aef\u7684\u5185\u5b58\u5206\u914d\u591a\u5c11 \u4e00\u4e2a Reduce Task \u53ef\u4f7f\u7528\u7684\u8d44\u6e90\u4e0a\u9650\uff08\u5355\u4f4d:MB\uff09\uff0c\u9ed8\u8ba4\u4e3a 1024\u3002\u5982\u679c Reduce Task \u5b9e\u9645\u4f7f\u7528\u7684\u8d44\u6e90\u91cf\u8d85\u8fc7\u8be5\u503c\uff0c\u5219\u4f1a\u88ab\u5f3a\u5236\u6740\u6b7b\n  &lt;property&gt;\n      &lt;name&gt;mapreduce.reduce.memory.mb&lt;\/name&gt;\n      &lt;value&gt;1280&lt;\/value&gt;\n&lt;\/property&gt;\n#  reduce\u7aef\u542f\u7528JVM\u6240\u5360\u7528\u7684\u5185\u5b58\u5927\u5c0f\n#java\u7684\u5806\u6808\u80af\u5b9a\u8981\u5c0f\u4e8ememory\u7684\uff0c\u56e0\u4e3a\u8fd8\u8981\u8003\u8651\u975e\u5806\u548c\u5176\u4ed6,\u4e0a\u9762\u6700\u5927\u80fd\u5206\u914d\u591a\u5c11\uff0c\u8fd8\u53d6\u51b3\u4e8eyarn\u7684\u914d\u7f6e\u3002\n  &lt;property&gt;\n      &lt;name&gt;Mapreduce.reduce.java.opts&lt;\/name&gt;\n      &lt;value&gt;-Xmx1024m -XX:-UseGCOverheadLimit&lt;\/value&gt;\n&lt;\/property&gt;\n#\u6bcf\u4e2a Maptask \u53ef\u7528\u7684\u6700\u591a cpu core \u6570\u76ee, \u9ed8\u8ba4\u503c: 1\n  &lt;property&gt;\n      &lt;name&gt;mapreduce.map.cpu.vcores&lt;\/name&gt;\n      &lt;value&gt;1&lt;\/value&gt;\n&lt;\/property&gt;\n#\u6bcf\u4e2a Reducetask \u53ef\u7528\u6700\u591a cpu core \u6570 \u9ed8\u8ba4\u503c: 1\n  &lt;property&gt;\n      &lt;name&gt; mapreduce.reduce.cpu.vcores&lt;\/name&gt;\n      &lt;value&gt;1&lt;\/value&gt;\n&lt;\/property&gt;\n<\/code><\/pre>\n<blockquote><p>\n  mapreduce shuffle \u4e0espark shuffle\u7684\u533a\u522b\n<\/p><\/blockquote>\n<p>\u7531\u4e8eShuffle\u6d89\u53ca\u5230\u4e86\u78c1\u76d8\u7684\u8bfb\u5199\u548c\u7f51\u7edc\u7684\u4f20\u8f93\uff0c\u56e0\u6b64Shuffle\u6027\u80fd\u7684\u9ad8\u4f4e\u76f4\u63a5\u5f71\u54cd\u5230\u4e86\u6574\u4e2a\u7a0b\u5e8f\u7684\u8fd0\u884c\u6548\u7387\u3002<br \/>\nspark\u7684shuffle\u5206\u4e3a\u4e24\u79cd\u5b9e\u73b0\uff0c\u5206\u522b\u4e3aHashShuffle\u548cSortShuffle\u3002<br \/>\n1. <code>HashShuffle<\/code><\/p>\n<p>\u53c8\u5206\u4e3a\u666e\u901a\u673a\u5236\u548c\u5408\u5e76\u673a\u5236\uff08consolidation\uff09\uff0c\u666e\u901a\u673a\u5236\u56e0\u4e3a\u5176\u4f1a\u4ea7\u751fM * R\u4e2a\u6570\u7684\u5de8\u91cf\u78c1\u76d8\u5c0f\u6587\u4ef6\u800c\u4ea7\u751f\u5927\u91cf\u6027\u80fd\u4f4e\u4e0b\u7684Io\u64cd\u4f5c\uff0c\u4ece\u800c\u6027\u80fd\u8f83\u4f4e\uff0c\u56e0\u4e3a\u5176\u5de8\u91cf\u7684\u78c1\u76d8\u5c0f\u6587\u4ef6\u8fd8\u53ef\u80fd\u5bfc\u81f4OOM\uff0cHashShuffle\u7684\u5408\u5e76\u673a\u5236\u901a\u8fc7\u91cd\u590d\u5229\u7528buffer\u7f13\u5b58\u4ece\u800c\u5c06\u78c1\u76d8\u5c0f\u6587\u4ef6\u7684\u6570\u91cf\u964d\u4f4e\u5230Core * R\u4e2a\uff0c\u4f46\u662f\u5f53Reducer \u7aef\u7684\u5e76\u884c\u4efb\u52a1\u6216\u8005\u662f\u6570\u636e\u5206\u7247\u8fc7\u591a\u7684\u65f6\u5019\uff0c\u4f9d\u7136\u4f1a\u4ea7\u751f\u5927\u91cf\u7684\u78c1\u76d8\u5c0f\u6587\u4ef6\u3002<\/p>\n<ol start=\"2\">\n<li><code>SortShuffle<\/code><\/li>\n<\/ol>\n<p>\u4e5f\u5206\u4e3a\u666e\u901a\u673a\u5236\u548cbypass\u673a\u5236\uff0c\u666e\u901a\u673a\u5236\u5728\u5185\u5b58\u6570\u636e\u7ed3\u6784(\u9ed8\u8ba4\u4e3a5M)\u5b8c\u6210\u6392\u5e8f\uff0c\u4f1a\u4ea7\u751f\u591a\u4e2a\u78c1\u76d8\u5c0f\u6587\u4ef6\uff0c\u6700\u540e\u5408\u5e76\u8fd9\u4e9b\u5c0f\u6587\u4ef6\u3002\u800c\u5f53shuffle map task\u6570\u91cf\u5c0f\u4e8espark.shuffle.sort.bypassMergeThreshold\u53c2\u6570\u7684\u503c\u3002\u6216\u8005\u7b97\u5b50\u4e0d\u662f\u805a\u5408\u7c7b(\u6bd4\u5982reduceByKey)\u7684shuffle\u7b97\u5b50\u7684\u65f6\u5019\u4f1a\u89e6\u53d1SortShuffle\u7684bypass\u673a\u5236\uff0cSortShuffle\u7684bypass\u673a\u5236\u4e0d\u4f1a\u8fdb\u884c\u6392\u5e8f\uff0c\u6781\u5927\u7684\u63d0\u9ad8\u4e86\u5176\u6027\u80fd\u3002<\/p>\n<ul>\n<li>\u533a\u522b<\/li>\n<\/ul>\n<p><code>\u5728MapReduce\u4e2d<\/code>\uff0cmap task\u5fc5\u987b\u5c06\u6240\u6709\u7684\u6570\u636e\u90fd\u5199\u5165\u672c\u5730\u78c1\u76d8\u6587\u4ef6\u4ee5\u540e\uff0c\u624d\u80fd\u542f\u52a8reduce\u64cd\u4f5c\uff0c\u6765\u62c9\u53d6\u6570\u636e\u3002\u4e3a\u4ec0\u4e48\uff1f\u56e0\u4e3amapreduce\u8981\u5b9e\u73b0\u9ed8\u8ba4\u7684\u6839\u636ekey\u7684\u6392\u5e8f\uff01\u6240\u4ee5\u8981\u6392\u5e8f\uff0c\u80af\u5b9a\u5f97\u5199\u5b8c\u6240\u6709\u6570\u636e\uff0c\u624d\u80fd\u6392\u5e8f\uff0c\u7136\u540ereduce\u6765\u62c9\u53d6\u3002<br \/>\n<code>\u4f46\u662fSpark\u4e0d\u9700\u8981<\/code>\uff0cspark\u9ed8\u8ba4\u60c5\u51b5\u4e0b\uff0c\u662f\u4e0d\u4f1a\u5bf9\u6570\u636e\u8fdb\u884c\u6392\u5e8f\u7684\u3002\u56e0\u6b64ShuffleMapTask\u6bcf\u5199\u5165\u4e00\u70b9\u6570\u636e\uff0cResultTask\u5c31\u53ef\u4ee5\u62c9\u53d6\u4e00\u70b9\u6570\u636e\uff0c\u7136\u540e\u5728\u672c\u5730\u6267\u884c\u6211\u4eec\u5b9a\u4e49\u7684\u805a\u5408\u51fd\u6570\u548c\u7b97\u5b50\uff0c\u8fdb\u884c\u8ba1\u7b97\u3002<br \/>\nspark\u8fd9\u79cd\u673a\u5236\u7684\u597d\u5904\u5728\u4e8e\uff0c\u901f\u5ea6\u6bd4mapreduce\u5feb\u591a\u4e86\u3002\u4f46\u662f\u4e5f\u6709\u4e00\u4e2a\u95ee\u9898\uff0cmapreduce\u63d0\u4f9b\u7684reduce\uff0c\u662f\u53ef\u4ee5\u5904\u7406\u6bcf\u4e2akey\u5bf9\u5e94\u7684value\u4e0a\u7684\uff0c\u5f88\u65b9\u4fbf\u3002\u4f46\u662fspark\u4e2d\uff0c\u7531\u4e8e\u8fd9\u79cd\u5b9e\u65f6\u62c9\u53d6\u7684\u673a\u5236\uff0c\u56e0\u6b64\u63d0\u4f9b\u4e0d\u4e86\u76f4\u63a5\u5904\u7406key\u5bf9\u5e94\u7684values\u7684\u7b97\u5b50\uff0c\u53ea\u80fd\u901a\u8fc7groupByKey\uff0c\u5148shuffle\uff0c\u4ea7\u751f\u4e00\u4e2aMapPartitionsRDD\uff0c\u7136\u540e\u7528map\u7b97\u5b50\uff0c\u6765\u5904\u7406\u6bcf\u4e2akey\u5bf9\u5e94\u7684values\uff0c\u5c31\u6ca1\u6709mapreduce\u7684\u8ba1\u7b97\u6a21\u578b\u90a3\u4e48\u65b9\u4fbf\u3002<\/p>\n<blockquote><p>\n  mapreduce\u600e\u4e48\u907f\u514d\u6570\u636e\u503e\u659c\n<\/p><\/blockquote>\n<p>\u662f\u4ec0\u4e48\uff1a\u6570\u636e\u503e\u659c\u5c31\u662f\u6570\u636e\u7684key \u7684\u5206\u5316\u4e25\u91cd\u4e0d\u5747\uff0c\u9020\u6210\u4e00\u90e8\u5206\u6570\u636e\u5f88\u591a\uff0c\u4e00\u90e8\u5206\u6570\u636e\u5f88\u5c11\u7684\u5c40\u9762<\/p>\n<p>\u81ea\u5df1\u5b9e\u73b0partition\u7c7b\uff0c\u7528key\u548cvalue\u76f8\u52a0\u53d6hash\u503c<\/p>\n<pre><code class=\"language-java line-numbers\">public int getPartition(K key, V value,int numReduceTasks) {\n\n    return ((\uff08key).hashCode()+value.hashCode()\uff09 &amp; Integer.MAX_VALUE) % numReduceTasks;\n\n  }\n<\/code><\/pre>\n<h1><span class=\"ez-toc-section\" id=\"%E8%A7%A3%E5%86%B3%E4%B8%80%E4%BA%9B%E6%8A%A5%E9%94%99\"><\/span>\u89e3\u51b3\u4e00\u4e9b\u62a5\u9519<span class=\"ez-toc-section-end\"><\/span><\/h1>\n<p>ideal\u64cd\u4f5cmapreduce\u62a5\u9519<br \/>\n<code>Could not locate executable null\\bin\\winutils.exe in the Hadoop binaries<\/code><a class=\"wp-editor-md-post-content-link\" href=\"https:\/\/github.com\/cdarlint\/winutils\/tree\/master\/hadoop-2.7.7\/bin\" title=\"\u70b9\u51fb\u8fdb\u5165\u4e0b\u8f7dwinnuts\">\u70b9\u51fb\u8fdb\u5165\u4e0b\u8f7dwinnuts<\/a><br \/>\n<code>Exceptionin thread \"main\" java.lang.UnsatisfiedLinkError:org.apache.hadoop.util.NativeCrc32.nativeCo<\/code><br \/>\n\u628a\u81ea\u5df1\u4e0b\u8f7d\u7684hadoop.dll\u590d\u5236\u5230<code>C:\\Windows\\System32<\/code><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u5b98\u7f51\u7ed9\u51fa\u7684\u4ecb\u7ecd Hadoop MapReduce is a software framework for ea [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[7],"tags":[],"_links":{"self":[{"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/posts\/1718"}],"collection":[{"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1718"}],"version-history":[{"count":5,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/posts\/1718\/revisions"}],"predecessor-version":[{"id":1920,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=\/wp\/v2\/posts\/1718\/revisions\/1920"}],"wp:attachment":[{"href":"http:\/\/www.specialwu.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1718"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1718"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.specialwu.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1718"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}