{"id":289,"date":"2024-06-14T10:01:00","date_gmt":"2024-06-14T02:01:00","guid":{"rendered":"https:\/\/solost.cn\/?p=289"},"modified":"2024-06-14T12:05:00","modified_gmt":"2024-06-14T04:05:00","slug":"424","status":"publish","type":"post","link":"https:\/\/solost.cn\/?p=289","title":{"rendered":"\u5927\u6570\u636e\u671f\u672b"},"content":{"rendered":"<h2>\u5927\u6570\u636e\u8ba1\u7b97\u6a21\u5f0f\u673a\u5668\u4ee3\u8868\u4f5c\u54c1<\/h2>\n<p><img decoding=\"async\" src=\"https:\/\/amarantos.oss-cn-hangzhou.aliyuncs.com\/typecho\/2024%2F06%2F14%2Fimage-20240614080939571.png\" alt=\"image-20240614080939571\" title=\"image-20240614080939571\"><\/p>\n<h2>\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u7269\u7406\u7ed3\u6784<\/h2>\n<p>\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u5728\u7269\u7406\u7ed3\u6784\u4e0a\u662f\u7531\u8ba1\u7b97\u673a\u96c6\u7fa4\u4e2d\u7684\u591a\u4e2a\u8282\u70b9\u6784\u6210\u7684\uff0c\u5982\u56fe3-2\u6240\u793a\u3002\u8fd9\u4e9b\u8282\u70b9\u5206\u4e3a\u4e24\u7c7b\uff1a<\/p>\n<p>\u4e3b\u8282\u70b9\uff08\u540d\u79f0\u8282\u70b9\uff09\uff1a\u6587\u4ef6\u76ee\u5f55\u7684\u521b\u5efa\u3001\u5220\u9664\u3001\u91cd\u547d\u540d\uff0c\u7ba1\u7406\u6570\u636e\u8282\u70b9\u548c\u6587\u4ef6\u5757\u7684\u6620\u5c04\u5173\u7cfb\u3002\u5ba2\u6237\u7aef\u53ea\u6709\u8bbf\u95ee\u540d\u79f0\u8282\u70b9\u624d\u80fd\u627e\u5230\u8bf7\u6c42\u7684\u6587\u4ef6\u5757\u6240\u5728\u7684\u4f4d\u7f6e\uff0c\u8fdb\u800c\u5230\u76f8\u5e94\u4f4d\u7f6e\u8bfb\u53d6\u6240\u9700\u6587\u4ef6\u5757\u3002<\/p>\n<p>\u4ece\u8282\u70b9\uff08\u6570\u636e\u8282\u70b9\uff09\uff1a\u8d1f\u8d23\u6570\u636e\u7684\u5b58\u50a8\u548c\u8bfb\u53d6\u3002\u5728\u5b58\u50a8\u65f6\uff0c\u7531\u540d\u79f0\u8282\u70b9\u5206\u914d\u5b58\u50a8\u4f4d\u7f6e\uff0c\u7136\u540e\u7531\u5ba2\u6237\u7aef\u628a\u6570\u636e\u76f4\u63a5\u5199\u5165\u76f8\u5e94\u6570\u636e\u8282\u70b9\uff1b\u5728\u8bfb\u53d6\u65f6\uff0c\u5ba2\u6237\u7aef\u4ece\u540d\u79f0\u8282\u70b9\u83b7\u5f97\u6570\u636e\u8282\u70b9\u548c\u6587\u4ef6\u5757\u7684\u6620\u5c04\u5173\u7cfb\uff0c\u5230\u76f8\u5e94\u4f4d\u7f6e\u8bbf\u95ee\u6587\u4ef6\u5757\u3002\u6570\u636e\u8282\u70b9\u4e5f\u8981\u6839\u636e\u540d\u79f0\u8282\u70b9\u7684\u547d\u4ee4\u521b\u5efa\u3001\u5220\u9664\u548c\u590d\u5236\u6570\u636e\u5757\u3002<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/amarantos.oss-cn-hangzhou.aliyuncs.com\/typecho\/2024%2F06%2F14%2Fimage-20240614081938181.png\" alt=\"image-20240614081938181\" title=\"image-20240614081938181\"><\/p>\n<h2>HDFS \u526f\u672c\u5b58\u653e\u7b56\u7565<\/h2>\n<p>(1)<\/p>\n<ul>\n<li>\u5728\u96c6\u7fa4\u5185\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\uff0c\u5219\u628a\u7b2c1\u4e2a\u526f\u672c\u653e\u7f6e\u5728\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\u7684\u6570\u636e\u8282\u70b9\u4e0a\uff0c\u5b9e\u73b0\u5c31\u8fd1\u5199\u5165\u6570\u636e\u3002<\/li>\n<li>\u5728\u96c6\u7fa4\u5916\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\uff0c\u5219\u4ece\u96c6\u7fa4\u5185\u90e8\u6311\u9009\u4e00\u53f0\u78c1\u76d8\u7a7a\u95f4\u8f83\u4e3a\u5145\u8db3\u3001CPU\u4e0d\u592a\u5fd9\u7684\u6570\u636e\u8282\u70b9\uff0c\u4f5c\u4e3a\u7b2c1\u4e2a\u526f\u672c\u7684\u5b58\u653e\u5730\u3002<\/li>\n<\/ul>\n<p>(2)\u7b2c2\u4e2a\u526f\u672c\u4f1a\u88ab\u653e\u7f6e\u5728\u4e0e\u7b2c1\u4e2a\u526f\u672c\u4e0d\u540c\u7684\u673a\u67b6\u7684\u6570\u636e\u8282\u70b9\u4e0a\u3002<br \/>(3)\u7b2c3\u4e2a\u526f\u672c\u4f1a\u88ab\u653e\u7f6e\u5728\u4e0e\u7b2c1\u4e2a\u526f\u672c\u76f8\u540c\u7684\u673a\u67b6\u7684\u5176\u4ed6\u8282\u70b9\u4e0a\u3002<br \/>(4)\u5982\u679c\u8fd8\u6709\u66f4\u591a\u7684\u526f\u672c\uff0c\u5219\u7ee7\u7eed\u4ece\u96c6\u7fa4\u4e2d\u968f\u673a\u9009\u62e9\u6570\u636e\u8282\u70b9\u8fdb\u884c\u5b58\u653e\u3002<\/p>\n<h2>Shuffle\u8fc7\u7a0b<\/h2>\n<p>(1)\u5728Map\u7aef\u7684Shuffle\u8fc7\u7a0b<br \/>Map\u4efb\u52a1\u7684\u8f93\u51fa\u7ed3\u679c\u9996\u5148\u88ab\u5199\u5165\u7f13\u5b58\uff0c\u5f53\u7f13\u5b58\u6ee1\u65f6\uff0c\u5c31\u542f\u52a8\u6ea2\u5199\u64cd\u4f5c\uff0c\u628a\u7f13\u5b58\u4e2d\u7684\u6570\u636e\u5199\u5165\u78c1\u76d8\u6587\u4ef6\uff0c\u5e76\u6e05\u7a7a\u7f13\u5b58\u3002\u5f53\u542f\u52a8\u6ea2\u5199\u64cd\u4f5c\u65f6\uff0c\u9996\u5148\u9700\u8981\u5bf9\u7f13\u5b58\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u5206\u533a\uff0c\u7136\u540e\u5bf9\u6bcf\u4e2a\u5206\u533a\u7684\u6570\u636e\u8fdb\u884c\u6392\u5e8f\u548c\u5408\u5e76\uff0c\u518d\u5199\u5165\u78c1\u76d8\u6587\u4ef6\u3002\u6bcf\u6b21\u6ea2\u5199\u64cd\u4f5c\u4f1a\u751f\u6210\u4e00\u4e2a\u65b0\u7684\u78c1\u76d8\u6587\u4ef6\uff0c\u968f\u7740Map\u4efb\u52a1\u7684\u6267\u884c\uff0c\u78c1\u76d8\u4e2d\u5c31\u4f1a\u751f\u6210\u591a\u4e2a\u6ea2\u5199\u6587\u4ef6\u3002\u5728Map\u4efb\u52a1\u5168\u90e8\u7ed3\u675f\u4e4b\u524d\uff0c\u8fd9\u4e9b\u6ea2\u5199\u6587\u4ef6\u4f1a\u88ab\u5f52\u5e76\u6210\u4e00\u4e2a\u5927\u7684\u78c1\u76d8\u6587\u4ef6\uff0c\u7136\u540e\u901a\u77e5\u76f8\u5e94\u7684Reduce\u4efb\u52a1\u6765\u201c\u9886\u53d6\u201d\u5c5e\u4e8e\u81ea\u5df1\u5904\u7406\u7684\u6570\u636e\u3002<\/p>\n<p>(2)\u5728Reduce\u7aef\u7684Shuffle\u8fc7\u7a0b<\/p>\n<p>Reduce\u4efb\u52a1\u4eceMap\u7aef\u7684\u4e0d\u540cMap\u673a\u5668\u201c\u9886\u53d6\u201d\u5c5e\u4e8e\u81ea\u5df1\u5904\u7406\u7684\u90a3\u90e8\u5206\u6570\u636e\uff0c\u7136\u540e\u5bf9\u6570\u636e\u8fdb\u884c\u5f52\u5e76\u540e\u4ea4\u7ed9Reduce\u5904\u7406\u3002<\/p>\n<h2>WordConunt\u8bbe\u8ba1\u601d\u8def<\/h2>\n<ol>\n<li><strong>\u5206\u7247\uff08Splitting\uff09\u9636\u6bb5<\/strong>\uff1a\u4e00\u4e2a\u5927\u6587\u4ef6\u88ab\u5207\u5206\u6210\u591a\u4e2a\u5206\u7247\uff0c\u5206\u914d\u5230\u591a\u4e2a\u4e0d\u540c\u7684\u673a\u5668\u4e0a\u3002\u6bcf\u4e2a\u673a\u5668\u4e0a\u7684Map\u4efb\u52a1\u8d1f\u8d23\u5904\u7406\u5206\u914d\u7ed9\u5b83\u7684\u90a3\u4e00\u90e8\u5206\u6570\u636e\u3002<\/li>\n<li><strong>Map\u9636\u6bb5<\/strong>\uff1a\u5728Map\u4efb\u52a1\u4e2d\uff0c\u8f93\u5165\u88ab\u9ed8\u8ba4\u5904\u7406\u4e3a\u952e\u503c\u5bf9\u5f62\u5f0f\uff0c\u5176\u4e2d\u952e\u662f\u6587\u4ef6\u7684\u884c\u53f7\uff0c\u503c\u662f\u5bf9\u5e94\u7684\u884c\u5185\u5bb9\u3002Map\u4efb\u52a1\u89e3\u6790\u6bcf\u884c\u7684\u5185\u5bb9\uff0c\u8bc6\u522b\u51fa\u5355\u8bcd\uff0c\u5e76\u8f93\u51fa\u952e\u503c\u5bf9\uff0c\u683c\u5f0f\u4e3a&lt;\u5355\u8bcd\uff0c1&gt;\uff0c\u8868\u793a\u8fd9\u4e2a\u5355\u8bcd\u51fa\u73b0\u4e00\u6b21\u3002<\/li>\n<li><strong>Shuffle\u9636\u6bb5<\/strong>\uff1aMap\u9636\u6bb5\u7684\u8f93\u51fa\u7ecf\u8fc7\u6392\u5e8f\u548c\u5206\u533a\u5904\u7406\uff0c\u8f6c\u6362\u6210&lt;key, value-list&gt;\u7684\u5f62\u5f0f\uff0c\u4f8b\u5982&lt;hadoop, [1, 1, 1, 1, 1]&gt;\u3002\u8fd9\u8868\u660e\u5355\u8bcd\"hadoop\"\u51fa\u73b0\u4e86\u4e94\u6b21\u3002<\/li>\n<li><strong>Reduce\u9636\u6bb5<\/strong>\uff1aReduce\u4efb\u52a1\u63a5\u6536\u6765\u81eaShuffle\u9636\u6bb5\u7684\u4e2d\u95f4\u7ed3\u679c\uff0c\u5f00\u59cb\u8fdb\u884c\u6c47\u603b\u8ba1\u7b97\u3002\u6bcf\u4e2aReduce\u4efb\u52a1\u8d1f\u8d23\u4e00\u90e8\u5206\u6570\u636e\u7684\u6c47\u603b\uff0c\u8ba1\u7b97\u6bcf\u4e2a\u5355\u8bcd\u7684\u6700\u7ec8\u9891\u6570\uff0c\u5e76\u5c06\u8fd9\u4e9b\u7ed3\u679c\u8f93\u51fa\u5230\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u4e2d\u3002<\/li>\n<\/ol>\n<p><img decoding=\"async\" src=\"https:\/\/amarantos.oss-cn-hangzhou.aliyuncs.com\/typecho\/2024%2F06%2F14%2Fimage-20240614081406011.png\" alt=\"image-20240614081406011\" title=\"image-20240614081406011\"><\/p>\n<h2>Spark\u4e3b\u8981\u5177\u6709\u5982\u4e0b\u4f18\u70b9\u3002<\/h2>\n<p>(l)Spark\u7684\u8ba1\u7b97\u6a21\u5f0f\u4e5f\u5c5e\u4e8eMapReduce,\u4f46\u4e0d\u5c40\u9650\u4e8eMap\u548cReduce\u64cd\u4f5c\uff0c\u8fd8\u63d0\u4f9b\u4e86\u591a\u79cd\u6570\u636e\u96c6\u64cd\u4f5c\u7c7b\u578b\uff0c\u7f16\u7a0b\u6a21\u578b\u6bd4MapReduce\u66f4\u7075\u6d3b\u3002<br \/>(2)Spak\u63d0\u4f9b\u4e86\u5185\u5b58\u8ba1\u7b97\uff0c\u4e2d\u95f4\u7ed3\u679c\u76f4\u63a5\u653e\u5230\u5185\u5b58\u4e2d\uff0c\u5e26\u6765\u4e86\u66f4\u9ad8\u7684\u8fed\u4ee3\u8fd0\u7b97\u6548\u7387\u3002<br \/>(3)Spark\u57fa\u4e8eDAG\u7684\u4efb\u52a1\u8c03\u5ea6\u6267\u884c\u673a\u5236\uff0c\u8981\u4f18\u4e8eMapReduce\u7684\u8fed\u4ee3\u6267\u884c\u673a\u5236\u3002<\/p>\n<h2>Spark\u5404\u79cd\u6982\u5ff5\u4e4b\u95f4\u7684\u76f8\u4e92\u5173\u7cfb<\/h2>\n<p>\u5728Spark\u4e2d\uff0c\u4e00\u4e2a\u5e94\u7528(Application)\u7531\u4e00\u4e2a\u4efb\u52a1\u63a7\u5236\u8282\u70b9(Driver)\u548c\u82e5\u5e72\u4e2a\u4f5c\u4e1a(Job)\u6784\u6210\uff0c\u4e00\u4e2a\u4f5c\u4e1a\u7531\u591a\u4e2a\u9636\u6bb5(Stage)\u6784\u6210\uff0c\u4e00\u4e2a\u9636\u6bb5\u7531\u591a\u4e2a\u4efb\u52a1(Task)\u7ec4\u6210\u3002\u5f53\u6267\u884c\u4e00\u4e2a\u5e94\u7528\u65f6\uff0c\u4efb\u52a1\u63a7\u5236\u8282\u70b9\u4f1a\u5411\u96c6\u7fa4\u7ba1\u7406\u5668(Cluster Manager)\u7533\u8bf7\u8d44\u6e90\uff0c\u542f\u52a8Executor,\u5e76\u5411Executor\u53d1\u9001\u5e94\u7528\u7a0b\u5e8f\u4ee3\u7801\u548c\u6587\u4ef6\uff0c\u7136\u540e\u5728Executor\u4e0a\u6267\u884c\u4efb\u52a1\uff0c\u8fd0\u884c\u7ed3\u675f\u540e\u6267\u884c\u7ed3\u679c\u4f1a\u8fd4\u56de\u7ed9\u4efb\u52a1\u63a7\u5236\u8282\u70b9\uff0c\u6216\u8005\u5199\u5230HDFS\u6216\u8005\u5176\u4ed6\u6570\u636e\u5e93\u4e2d\u3002<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/amarantos.oss-cn-hangzhou.aliyuncs.com\/typecho\/2024%2F06%2F14%2Fimage-20240614081554772.png\" alt=\"image-20240614081554772\" title=\"image-20240614081554772\"><\/p>\n<h2>\u6d41\u8ba1\u7b97\u4e0eHadoop<\/h2>\n<p>Hadoop\u4f5c\u4e3a\u5927\u6570\u636e\u6280\u672f\u7684\u4e8b\u5b9e\u6807\u51c6\uff0c\u4e3b\u8981\u4f9d\u9760MapReduce\u548cHDFS\u6765\u652f\u6491\u5927\u89c4\u6a21\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u548c\u5904\u7406\u3002\u7136\u800c\uff0cHadoop\u7684MapReduce\u6a21\u578b\u4e3b\u8981\u9002\u7528\u4e8e\u6279\u91cf\u6570\u636e\u5904\u7406\uff0c\u4e0d\u9002\u5408\u6d41\u8ba1\u7b97\u7684\u5b9e\u65f6\u9700\u6c42\u3002\u5c3d\u7ba1\u53ef\u4ee5\u901a\u8fc7\u5c0f\u6279\u91cf\u5904\u7406\u6765\u51cf\u5c11\u5ef6\u8fdf\uff0c\u8fd9\u79cd\u65b9\u6cd5\u4f1a\u589e\u52a0\u5904\u7406\u5f00\u9500\u548c\u590d\u6742\u5ea6\uff0c\u964d\u4f4e\u7cfb\u7edf\u7684\u53ef\u7ef4\u62a4\u6027\u548c\u7528\u6237\u7a0b\u5e8f\u7684\u53ef\u4f38\u7f29\u6027\u3002\u56e0\u6b64\uff0c\u4e1a\u754c\u5df2\u7ecf\u5f00\u53d1\u4e86\u8bb8\u591a\u4e13\u95e8\u7684\u5b9e\u65f6\u6d41\u6570\u636e\u5904\u7406\u7cfb\u7edf\u6765\u6ee1\u8db3\u8fd9\u4e9b\u9700\u6c42\u3002<\/p>\n<h2>1\u3001\u7f16\u7a0b\u5b9e\u73b0\u6587\u4ef6\u5408\u5e76\u548c\u53bb\u91cd\u64cd\u4f5c<\/h2>\n<p>\u5bf9\u4e8e\u4e24\u4e2a\u8f93\u5165\u6587\u4ef6\uff0c\u5373\u6587\u4ef6 A \u548c\u6587\u4ef6 B\uff0c\u8bf7\u7f16\u5199 MapReduce \u7a0b\u5e8f\uff0c\u5bf9\u4e24\u4e2a\u6587\u4ef6\u8fdb\u884c\u5408\u5e76\uff0c\u5e76\u5254\u9664\u5176\u4e2d\u91cd\u590d\u7684\u5185\u5bb9\uff0c\u5f97\u5230\u4e00\u4e2a\u65b0\u7684\u8f93\u51fa\u6587\u4ef6 C\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u6587\u4ef6\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002<\/p>\n<p>\u8f93\u5165\u6587\u4ef6A\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 x\n20150102 y\n20150103 x\n20150104 y\n20150105 z\n20150106 x<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6B\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 y\n20150102 y\n20150103 x\n20150104 Z\n20150105 y<\/code><\/pre>\n<p>\u6839\u636e\u8f93\u5165\u6587\u4ef6A\u548cB\u5408\u5e76\u5f97\u5230\u7684\u8f93\u51fa\u6587\u4ef6C\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 x\n20150101 y\n20150102 y\n20150103 x\n20150104 y\n20150104 z\n20150105 y\n20150105 z\n20150106 x<\/code><\/pre>\n<p>\u53c2\u8003\u4ee3\u7801<\/p>\n<pre><code class=\"lang-java\">public class MyTest {\n\n    public static class Map extends Mapper&lt;Object, Text, Text, Text&gt; {\n        private static Text text = new Text();\n\n        public void map(Object key, Text value, Context context) {\n            text = value;\n            context.write(text, new Text(&quot;&quot;));\n        }\n    }\n\n    public static class Reduce extends Reducer&lt;Text, Text, Text, Text&gt; {\n        public void reduce(Text key, Iterable&lt;Text&gt; values, Context context) {\n            context.write(key, new Text(&quot;&quot;));\n        }\n    }\n\n}<\/code><\/pre>\n<h2>2\u3001\u7f16\u5199MapReduce\u7a0b\u5e8f\u5b9e\u73b0\u5bf9\u8f93\u5165\u6587\u4ef6\u7684\u6392\u5e8f<\/h2>\n<p>\u73b0\u5728\u6709\u591a\u4e2a\u8f93\u5165\u6587\u4ef6\uff0c\u6bcf\u4e2a\u6587\u4ef6\u4e2d\u7684\u6bcf\u884c\u5185\u5bb9\u5747\u4e3a\u4e00\u4e2a\u6574\u6570\u3002\u8981\u6c42\u8bfb\u53d6\u6240\u6709\u6587\u4ef6\u4e2d\u7684\u6574\u6570\uff0c\u8fdb\u884c\u5347\u5e8f\u6392\u5e8f\u540e\uff0c\u8f93\u51fa\u5230\u4e00\u4e2a\u65b0\u7684\u6587\u4ef6\u4e2d\uff0c\u8f93\u51fa\u7684\u6570\u636e\u683c\u5f0f\u4e3a\u6bcf\u884c\u4e24\u4e2a\u6574\u6570\uff0c\u7b2c\u4e00\u4e2a\u6570\u5b57\u4e3a\u7b2c\u4e8c\u4e2a\u6574\u6570\u7684\u6392\u5e8f\u4f4d\u6b21\uff0c\u7b2c\u4e8c\u4e2a\u6574\u6570\u4e3a\u539f\u5f85\u6392\u5217\u7684\u6574\u6570\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u6587\u4ef6\u7684\u4e00\u4e2a\u6837\u4f8b\u4f9b\u53c2\u8003\u3002<\/p>\n<p>\u8f93\u5165\u6587\u4ef6 1 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>33\n37\n12\n40<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6 2 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>4\n16\n39\n5<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6 3 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>1\n45\n25<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<pre><code>1 1\n2 4\n3 5\n4 12\n5 16\n6 25\n7 33\n8 37\n9 39\n10 40\n11 45<\/code><\/pre>\n<pre><code class=\"lang-java\">public class MyTest {\n    public static class Map extends Mapper&lt;Object, Text, IntWritable, IntWritable&gt; {\n        private static IntWritable data = new IntWritable();\n\n        public void map(Object key, Text value, Context context) {\n            String text = value.toString();\n            data.set(Integer.parseInt(text));\n            context.write(data, new IntWritable(1));\n        }\n    }\n\n    public static class Reduce extends Reducer&lt;IntWritable, IntWritable, IntWritable, IntWritable&gt; {\n        private static IntWritable line_num = new IntWritable(1);\n\n        public void reduce(IntWritable key, Iterable&lt;IntWritable&gt; values, Context context) {\n            for (IntWritable val : values) {\n                context.write(line_num, key);\n                line_num = new IntWritable(line_num.get() + 1);\n            }\n        }\n    }\n}<\/code><\/pre>\n<h2>3\u3001\u7f16\u5199 MapReduce \u7a0b\u5e8f\u5b9e\u73b0\u6c42\u5e73\u5747\u503c<\/h2>\n<p>\u6bcf\u4e2a\u8f93\u5165\u6587\u4ef6\u8868\u793a\u5b66\u751f\u67d0\u95e8\u8bfe\u7a0b\u7684\u6210\u7ee9\uff0c\u8f93\u5165\u6587\u4ef6\u4e2d\u6bcf\u884c\u5185\u5bb9\u7531\u4e24\u4e2a\u5b57\u6bb5\u7ec4\u6210\uff0c\u7b2c\u4e00\u4e2a\u5b57\u6bb5\u662f\u5b66\u751f\u59d3\u540d\uff0c\u7b2c\u4e8c\u4e2a\u5b57\u6bb5\u662f\u5b66\u751f\u7684\u6210\u7ee9\uff1b\u7f16\u5199 MapReduce \u5e94\u7528\u7a0b\u5e8f\u6c42\u51fa\u6240\u6709\u5b66\u751f\u7684\u5e73\u5747\u6210\u7ee9\uff0c\u5e76\u8f93\u51fa\u5230\u4e00\u4e2a\u65b0\u6587\u4ef6\u4e2d\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u6587\u4ef6\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002<\/p>\n<p>\u8f93\u5165\u6587\u4ef6(AlgorithmScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 92\nXiaoHong 87\nXiaoXin 82\nXiaoLi 90<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6(DataBaseScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 95\nXiaoHong 81\nXiaoXin 89\nXiaoLi 85<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6(PythonScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 82\nXiaoHong 83\nXiaoXin 94\nXiaoLi 91<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoHong 83.67\nXiaoLi 88.67\nXiaoMing 89.67\nXiaoXin 88.33<\/code><\/pre>\n<pre><code class=\"lang-java\">public class MyTest {\n\n    public static class MyMapper extends Mapper&lt;Object, Text, Text, FloatWritable&gt; {\n        private Text stuName = new Text();\n        private FloatWritable stuScore = new FloatWritable();\n\n        public void map(Object key, Text value, Context context) {\n            String[] line = value.toString().split(&quot; &quot;);\n            this.stuName.set(line[0]);\n            this.stuScore.set(Float.parseFloat(line[1]));\n            context.write(this.stuName, this.stuScore);\n        }\n    }\n\n    public static class MyReducer extends Reducer&lt;Text, FloatWritable, Text, FloatWritable&gt; {\n        private FloatWritable avgScore = new FloatWritable();\n\n        public void reduce(Text key, Iterable&lt;FloatWritable&gt; values, Context context) {\n            int num = 0;\n            float sum = 0;\n            float avg = 0;\n            for (FloatWritable val : values) {\n                sum += val.get();\n                num++;\n            }\n            avg = sum \/ num;\n            BigDecimal b = new BigDecimal(avg);\n            avgScore.set(b.setScale(2, BigDecimal.ROUND_HALF_UP).floatValue());\n            context.write(key, avgScore);\n        }\n    }\n}<\/code><\/pre>\n<h2>1\u3001\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\u5b9e\u73b0\u6570\u636e\u53bb\u91cd<\/h2>\n<p>\u5bf9\u4e8e\u4e24\u4e2a\u8f93\u5165\u6587\u4ef6 A \u548c B\uff0c\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\uff0c\u5bf9\u4e24\u4e2a\u6587\u4ef6\u8fdb\u884c\u5408\u5e76\u3001\u5254\u9664\u5176\u4e2d\u91cd\u590d\u7684\u5185\u5bb9\u5e76\u6392\u5e8f\uff0c\u5e76\u8f93\u51fa\u5230\u7ec8\u7aef\u4e2d\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u7ed3\u679c\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002<br \/>\u8f93\u5165\u6587\u4ef6A\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 x\n20150102 y\n20150103 x\n20150104 y\n20150105 z\n20150106 x<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6B\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 y\n20150102 y\n20150103 x\n20150104 Z\n20150105 y<\/code><\/pre>\n<p>\u6839\u636e\u8f93\u5165\u6587\u4ef6A\u548cB\u5408\u5e76\u5f97\u5230\u7684\u8f93\u51fa\u6587\u4ef6C\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>20150101 x\n20150101 y\n20150102 y\n20150103 x\n20150104 y\n20150104 z\n20150105 y\n20150105 z\n20150106 x<\/code><\/pre>\n<pre><code class=\"lang-scala\">object MyTest {\n  def main(args: Array[String]): Unit = {\n    val conf = new SparkConf().setAppName(&quot;MyTest&quot;).setMaster(&quot;local&quot;)\n    val sc = new SparkContext(conf)\n    val data = sc.textFile(&quot;file:\/\/\/home\/hadoop\/myfile&quot;)\n      \n    val rdd1 = data.map(line =&gt; (line, &quot;&quot;))\n    val rdd2 = rdd1.groupByKey()\n    val rdd3 = rdd2.sortByKey()\n    val rdd4 = rdd3.keys\n\n    rdd4.foreach(println)\n  }\n}<\/code><\/pre>\n<h2>2\u3001\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\u5b9e\u73b0\u6574\u5408\u6392\u5e8f<\/h2>\n<p>\u5bf9\u4e8e\u4e24\u4e2a\u8f93\u5165\u6587\u4ef6 A \u548c B\uff0c\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\uff0c\u5bf9\u4e24\u4e2a\u6587\u4ef6\u8fdb\u884c\u5408\u5e76\u3001\u5254\u9664\u5176\u4e2d\u91cd\u590d\u7684\u5185\u5bb9\u5e76\u6392\u5e8f\uff0c\u5e76\u8f93\u51fa\u5230\u7ec8\u7aef\u4e2d\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u7ed3\u679c\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002<\/p>\n<p>\u8f93\u5165\u6587\u4ef6 1 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>33\n37\n12\n40<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6 2 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>4\n16\n39\n5<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6 3 \u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>1\n45\n25<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<pre><code>1 1\n2 4\n3 5\n4 12\n5 16\n6 25\n7 33\n8 37\n9 39\n10 40\n11 45<\/code><\/pre>\n<pre><code class=\"lang-scala\">object MyTest {\n  def main(args: Array[String]): Unit = {\n    val conf = new SparkConf().setAppName(&quot;MyTest&quot;).setMaster(&quot;local&quot;)\n    val sc = new SparkContext(conf)\n    val data = sc.textFile(&quot;file:\/\/\/home\/hadoop\/myfile&quot;, 1)\n      \n    val rdd1 = data.map(line =&gt; (line.toInt, &quot;&quot;))\n    val rdd2 = rdd1.partitionBy(new HashPartitioner(1))\n    val rdd3 = rdd2.sortByKey()\n    val rdd4 = rdd3.keys\n\n    var index = 0\n    val rdd5 = rdd4.map(t =&gt; {\n      index = index + 1\n      (index, t)\n    })\n\n    rdd5.foreach(t =&gt; println(t._1 + &quot; &quot; + t._2))\n  }\n}<\/code><\/pre>\n<h2>3\u3001\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\u5b9e\u73b0\u6c42\u5e73\u5747\u503c<\/h2>\n<p>\u6bcf\u4e2a\u8f93\u5165\u6587\u4ef6\u8868\u793a\u5b66\u751f\u67d0\u95e8\u8bfe\u7a0b\u7684\u6210\u7ee9\uff0c\u8f93\u5165\u6587\u4ef6\u4e2d\u6bcf\u884c\u5185\u5bb9\u7531\u4e24\u4e2a\u5b57\u6bb5\u7ec4\u6210\uff0c\u7b2c\u4e00\u4e2a\u5b57\u6bb5\u662f\u5b66\u751f\u59d3\u540d\uff0c\u7b2c\u4e8c\u4e2a\u5b57\u6bb5\u662f\u5b66\u751f\u7684\u6210\u7ee9\uff1b\u7f16\u5199 Spark \u72ec\u7acb\u5e94\u7528\u7a0b\u5e8f\u6c42\u51fa\u6240\u6709\u5b66\u751f\u7684\u5e73\u5747\u6210\u7ee9\uff0c\u5e76\u8f93\u51fa\u5230\u7ec8\u7aef\u4e2d\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u7ed3\u679c\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002<br \/>\u8f93\u5165\u6587\u4ef6(AlgorithmScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 92\nXiaoHong 87\nXiaoXin 82\nXiaoLi 90<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6(DataBaseScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 95\nXiaoHong 81\nXiaoXin 89\nXiaoLi 85<\/code><\/pre>\n<p>\u8f93\u5165\u6587\u4ef6(PythonScore)\u7684\u6837\u4f8b\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoMing 82\nXiaoHong 83\nXiaoXin 94\nXiaoLi 91<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\u5982\u4e0b\uff1a<\/p>\n<pre><code>XiaoHong 83.67\nXiaoLi 88.67\nXiaoMing 89.67\nXiaoXin 88.33<\/code><\/pre>\n<pre><code class=\"lang-scala\">object MyTest {\n  def main(args: Array[String]): Unit = {\n    val conf = new SparkConf().setAppName(&quot;MyTest&quot;).setMaster(&quot;local&quot;)\n    val sc = new SparkContext(conf)\n    val data = sc.textFile(&quot;file:\/\/\/home\/hadoop\/myfile&quot;)\n      \n    val rdd1 = data.map(line =&gt; line.split(&quot; &quot;))\n    val rdd2 = rdd1.map(t =&gt; (t(0), t(1).toInt))\n    val rdd3 = rdd2.mapValues(x =&gt; (x, 1))\n    val rdd4 = rdd3.reduceByKey((x, y) =&gt; (x._1 + y._1, x._2 + y._2))\n    val rdd5 = rdd4.mapValues(x =&gt; x._1.toDouble \/ x._2)\n\n    rdd5.foreach(t =&gt; {\n      val x = t._2\n      println(t._1 + &quot; &quot; + f&quot;$x%1.2f&quot;)\n    })\n  }\n}<\/code><\/pre>\n","protected":false},"excerpt":{"rendered":"<p>\u5927\u6570\u636e\u8ba1\u7b97\u6a21\u5f0f\u673a\u5668\u4ee3\u8868\u4f5c\u54c1 \u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u7269\u7406\u7ed3\u6784 \u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u5728\u7269\u7406\u7ed3\u6784\u4e0a\u662f\u7531\u8ba1\u7b97\u673a\u96c6\u7fa4\u4e2d\u7684\u591a\u4e2a\u8282\u70b9\u6784\u6210\u7684\uff0c\u5982\u56fe3-2\u6240\u793a\u3002\u8fd9\u4e9b\u8282\u70b9\u5206\u4e3a\u4e24\u7c7b\uff1a \u4e3b\u8282\u70b9\uff08\u540d\u79f0\u8282\u70b9\uff09\uff1a\u6587\u4ef6\u76ee\u5f55\u7684\u521b\u5efa\u3001\u5220\u9664\u3001\u91cd\u547d\u540d\uff0c\u7ba1\u7406\u6570\u636e\u8282\u70b9\u548c\u6587\u4ef6\u5757\u7684\u6620\u5c04\u5173\u7cfb\u3002\u5ba2\u6237\u7aef\u53ea\u6709\u8bbf\u95ee\u540d\u79f0\u8282\u70b9\u624d\u80fd\u627e\u5230\u8bf7\u6c42\u7684\u6587\u4ef6\u5757\u6240\u5728\u7684\u4f4d\u7f6e\uff0c\u8fdb\u800c\u5230\u76f8\u5e94\u4f4d\u7f6e\u8bfb\u53d6\u6240\u9700\u6587\u4ef6\u5757\u3002 \u4ece\u8282\u70b9\uff08\u6570\u636e\u8282\u70b9\uff09\uff1a\u8d1f\u8d23\u6570\u636e\u7684\u5b58\u50a8\u548c\u8bfb\u53d6\u3002\u5728\u5b58\u50a8\u65f6\uff0c\u7531\u540d\u79f0\u8282\u70b9\u5206\u914d\u5b58\u50a8\u4f4d\u7f6e\uff0c\u7136\u540e\u7531\u5ba2\u6237\u7aef\u628a\u6570\u636e\u76f4\u63a5\u5199\u5165\u76f8\u5e94\u6570\u636e\u8282\u70b9\uff1b\u5728\u8bfb\u53d6\u65f6\uff0c\u5ba2\u6237\u7aef\u4ece\u540d\u79f0\u8282\u70b9\u83b7\u5f97\u6570\u636e\u8282\u70b9\u548c\u6587\u4ef6\u5757\u7684\u6620\u5c04\u5173\u7cfb\uff0c\u5230\u76f8\u5e94\u4f4d\u7f6e\u8bbf\u95ee\u6587\u4ef6\u5757\u3002\u6570\u636e\u8282\u70b9\u4e5f\u8981\u6839\u636e\u540d\u79f0\u8282\u70b9\u7684\u547d\u4ee4\u521b\u5efa\u3001\u5220\u9664\u548c\u590d\u5236\u6570\u636e\u5757\u3002 HDFS \u526f\u672c\u5b58\u653e\u7b56\u7565 (1) \u5728\u96c6\u7fa4\u5185\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\uff0c\u5219\u628a\u7b2c1\u4e2a\u526f\u672c\u653e\u7f6e\u5728\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\u7684\u6570\u636e\u8282\u70b9\u4e0a\uff0c\u5b9e\u73b0\u5c31\u8fd1\u5199\u5165\u6570\u636e\u3002 \u5728\u96c6\u7fa4\u5916\u53d1\u8d77\u5199\u64cd\u4f5c\u8bf7\u6c42\uff0c\u5219\u4ece\u96c6\u7fa4\u5185\u90e8\u6311\u9009\u4e00\u53f0\u78c1\u76d8\u7a7a\u95f4\u8f83\u4e3a\u5145\u8db3\u3001CPU\u4e0d\u592a\u5fd9\u7684\u6570\u636e\u8282\u70b9\uff0c\u4f5c\u4e3a\u7b2c1\u4e2a\u526f\u672c\u7684\u5b58\u653e\u5730\u3002 (2)\u7b2c2\u4e2a\u526f\u672c\u4f1a\u88ab\u653e\u7f6e\u5728\u4e0e\u7b2c1\u4e2a\u526f\u672c\u4e0d\u540c\u7684\u673a\u67b6\u7684\u6570\u636e\u8282\u70b9\u4e0a\u3002(3)\u7b2c3\u4e2a\u526f\u672c\u4f1a\u88ab\u653e\u7f6e\u5728\u4e0e\u7b2c1\u4e2a\u526f\u672c\u76f8\u540c\u7684\u673a\u67b6\u7684\u5176\u4ed6\u8282\u70b9\u4e0a\u3002(4)\u5982\u679c\u8fd8\u6709\u66f4\u591a\u7684\u526f\u672c\uff0c\u5219\u7ee7\u7eed\u4ece\u96c6\u7fa4\u4e2d\u968f\u673a\u9009\u62e9\u6570\u636e\u8282\u70b9\u8fdb\u884c\u5b58\u653e\u3002 Shuffle\u8fc7\u7a0b (1)\u5728Map\u7aef\u7684Shuffle\u8fc7\u7a0bMap\u4efb\u52a1\u7684\u8f93\u51fa\u7ed3\u679c\u9996\u5148\u88ab\u5199\u5165\u7f13\u5b58\uff0c\u5f53\u7f13\u5b58\u6ee1\u65f6\uff0c\u5c31\u542f\u52a8\u6ea2\u5199\u64cd\u4f5c\uff0c\u628a\u7f13\u5b58\u4e2d\u7684\u6570\u636e\u5199\u5165\u78c1\u76d8\u6587\u4ef6\uff0c\u5e76\u6e05\u7a7a\u7f13\u5b58\u3002\u5f53\u542f\u52a8\u6ea2\u5199\u64cd\u4f5c\u65f6\uff0c\u9996\u5148\u9700\u8981\u5bf9\u7f13\u5b58\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u5206\u533a\uff0c\u7136\u540e\u5bf9\u6bcf\u4e2a\u5206\u533a\u7684\u6570\u636e\u8fdb\u884c\u6392\u5e8f\u548c\u5408\u5e76\uff0c\u518d\u5199\u5165\u78c1\u76d8\u6587\u4ef6\u3002\u6bcf\u6b21\u6ea2\u5199\u64cd\u4f5c\u4f1a\u751f\u6210\u4e00\u4e2a\u65b0\u7684\u78c1\u76d8\u6587\u4ef6\uff0c\u968f\u7740Map\u4efb\u52a1\u7684\u6267\u884c\uff0c\u78c1\u76d8\u4e2d\u5c31\u4f1a\u751f\u6210\u591a\u4e2a\u6ea2\u5199\u6587\u4ef6\u3002\u5728Map\u4efb\u52a1\u5168\u90e8\u7ed3\u675f\u4e4b\u524d\uff0c\u8fd9\u4e9b\u6ea2\u5199\u6587\u4ef6\u4f1a\u88ab\u5f52\u5e76\u6210\u4e00\u4e2a\u5927\u7684\u78c1\u76d8\u6587\u4ef6\uff0c\u7136\u540e\u901a\u77e5\u76f8\u5e94\u7684Reduce\u4efb\u52a1\u6765\u201c\u9886\u53d6\u201d\u5c5e\u4e8e\u81ea\u5df1\u5904\u7406\u7684\u6570\u636e\u3002 (2)\u5728Reduce\u7aef\u7684Shuffle\u8fc7\u7a0b Reduce\u4efb\u52a1\u4eceMap\u7aef\u7684\u4e0d\u540cMap\u673a\u5668\u201c\u9886\u53d6\u201d\u5c5e\u4e8e\u81ea\u5df1\u5904\u7406\u7684\u90a3\u90e8\u5206\u6570\u636e\uff0c\u7136\u540e\u5bf9\u6570\u636e\u8fdb\u884c\u5f52\u5e76\u540e\u4ea4\u7ed9Reduce\u5904\u7406\u3002 WordConunt\u8bbe\u8ba1\u601d\u8def \u5206\u7247\uff08Splitting\uff09\u9636\u6bb5\uff1a\u4e00\u4e2a\u5927\u6587\u4ef6\u88ab\u5207\u5206\u6210\u591a\u4e2a\u5206\u7247\uff0c\u5206\u914d\u5230\u591a\u4e2a\u4e0d\u540c\u7684\u673a\u5668\u4e0a\u3002\u6bcf\u4e2a\u673a\u5668\u4e0a\u7684Map\u4efb\u52a1\u8d1f\u8d23\u5904\u7406\u5206\u914d\u7ed9\u5b83\u7684\u90a3\u4e00\u90e8\u5206\u6570\u636e\u3002 Map\u9636\u6bb5\uff1a\u5728Map\u4efb\u52a1\u4e2d\uff0c\u8f93\u5165\u88ab\u9ed8\u8ba4\u5904\u7406\u4e3a\u952e\u503c\u5bf9\u5f62\u5f0f\uff0c\u5176\u4e2d\u952e\u662f\u6587\u4ef6\u7684\u884c\u53f7\uff0c\u503c\u662f\u5bf9\u5e94\u7684\u884c\u5185\u5bb9\u3002Map\u4efb\u52a1\u89e3\u6790\u6bcf\u884c\u7684\u5185\u5bb9\uff0c\u8bc6\u522b\u51fa\u5355\u8bcd\uff0c\u5e76\u8f93\u51fa\u952e\u503c\u5bf9\uff0c\u683c\u5f0f\u4e3a&lt;\u5355\u8bcd\uff0c1&gt;\uff0c\u8868\u793a\u8fd9\u4e2a\u5355\u8bcd\u51fa\u73b0\u4e00\u6b21\u3002 Shuffle\u9636\u6bb5\uff1aMap\u9636\u6bb5\u7684\u8f93\u51fa\u7ecf\u8fc7\u6392\u5e8f\u548c\u5206\u533a\u5904\u7406\uff0c\u8f6c\u6362\u6210&lt;key, value-list&gt;\u7684\u5f62\u5f0f\uff0c\u4f8b\u5982&lt;hadoop, [1, 1, 1, 1, 1]&gt;\u3002\u8fd9\u8868\u660e\u5355\u8bcd\"hadoop\"\u51fa\u73b0\u4e86\u4e94\u6b21\u3002 Reduce\u9636\u6bb5\uff1aReduce\u4efb\u52a1\u63a5\u6536\u6765\u81eaShuffle\u9636\u6bb5\u7684\u4e2d\u95f4\u7ed3\u679c\uff0c\u5f00\u59cb\u8fdb\u884c\u6c47\u603b\u8ba1\u7b97\u3002\u6bcf\u4e2aReduce\u4efb\u52a1\u8d1f\u8d23\u4e00\u90e8\u5206\u6570\u636e\u7684\u6c47\u603b\uff0c\u8ba1\u7b97\u6bcf\u4e2a\u5355\u8bcd\u7684\u6700\u7ec8\u9891\u6570\uff0c\u5e76\u5c06\u8fd9\u4e9b\u7ed3\u679c\u8f93\u51fa\u5230\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u4e2d\u3002 Spark\u4e3b\u8981\u5177\u6709\u5982\u4e0b\u4f18\u70b9\u3002 (l)Spark\u7684\u8ba1\u7b97\u6a21\u5f0f\u4e5f\u5c5e\u4e8eMapReduce,\u4f46\u4e0d\u5c40\u9650\u4e8eMap\u548cReduce\u64cd\u4f5c\uff0c\u8fd8\u63d0\u4f9b\u4e86\u591a\u79cd\u6570\u636e\u96c6\u64cd\u4f5c\u7c7b\u578b\uff0c\u7f16\u7a0b\u6a21\u578b\u6bd4MapReduce\u66f4\u7075\u6d3b\u3002(2)Spak\u63d0\u4f9b\u4e86\u5185\u5b58\u8ba1\u7b97\uff0c\u4e2d\u95f4\u7ed3\u679c\u76f4\u63a5\u653e\u5230\u5185\u5b58\u4e2d\uff0c\u5e26\u6765\u4e86\u66f4\u9ad8\u7684\u8fed\u4ee3\u8fd0\u7b97\u6548\u7387\u3002(3)Spark\u57fa\u4e8eDAG\u7684\u4efb\u52a1\u8c03\u5ea6\u6267\u884c\u673a\u5236\uff0c\u8981\u4f18\u4e8eMapReduce\u7684\u8fed\u4ee3\u6267\u884c\u673a\u5236\u3002 Spark\u5404\u79cd\u6982\u5ff5\u4e4b\u95f4\u7684\u76f8\u4e92\u5173\u7cfb \u5728Spark\u4e2d\uff0c\u4e00\u4e2a\u5e94\u7528(Application)\u7531\u4e00\u4e2a\u4efb\u52a1\u63a7\u5236\u8282\u70b9(Driver)\u548c\u82e5\u5e72\u4e2a\u4f5c\u4e1a(Job)\u6784\u6210\uff0c\u4e00\u4e2a\u4f5c\u4e1a\u7531\u591a\u4e2a\u9636\u6bb5(Stage)\u6784\u6210\uff0c\u4e00\u4e2a\u9636\u6bb5\u7531\u591a\u4e2a\u4efb\u52a1(Task)\u7ec4\u6210\u3002\u5f53\u6267\u884c\u4e00\u4e2a\u5e94\u7528\u65f6\uff0c\u4efb\u52a1\u63a7\u5236\u8282\u70b9\u4f1a\u5411\u96c6\u7fa4\u7ba1\u7406\u5668(Cluster Manager)\u7533\u8bf7\u8d44\u6e90\uff0c\u542f\u52a8Executor,\u5e76\u5411Executor\u53d1\u9001\u5e94\u7528\u7a0b\u5e8f\u4ee3\u7801\u548c\u6587\u4ef6\uff0c\u7136\u540e\u5728Executor\u4e0a\u6267\u884c\u4efb\u52a1\uff0c\u8fd0\u884c\u7ed3\u675f\u540e\u6267\u884c\u7ed3\u679c\u4f1a\u8fd4\u56de\u7ed9\u4efb\u52a1\u63a7\u5236\u8282\u70b9\uff0c\u6216\u8005\u5199\u5230HDFS\u6216\u8005\u5176\u4ed6\u6570\u636e\u5e93\u4e2d\u3002 \u6d41\u8ba1\u7b97\u4e0eHadoop Hadoop\u4f5c\u4e3a\u5927\u6570\u636e\u6280\u672f\u7684\u4e8b\u5b9e\u6807\u51c6\uff0c\u4e3b\u8981\u4f9d\u9760MapReduce\u548cHDFS\u6765\u652f\u6491\u5927\u89c4\u6a21\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u548c\u5904\u7406\u3002\u7136\u800c\uff0cHadoop\u7684MapReduce\u6a21\u578b\u4e3b\u8981\u9002\u7528\u4e8e\u6279\u91cf\u6570\u636e\u5904\u7406\uff0c\u4e0d\u9002\u5408\u6d41\u8ba1\u7b97\u7684\u5b9e\u65f6\u9700\u6c42\u3002\u5c3d\u7ba1\u53ef\u4ee5\u901a\u8fc7\u5c0f\u6279\u91cf\u5904\u7406\u6765\u51cf\u5c11\u5ef6\u8fdf\uff0c\u8fd9\u79cd\u65b9\u6cd5\u4f1a\u589e\u52a0\u5904\u7406\u5f00\u9500\u548c\u590d\u6742\u5ea6\uff0c\u964d\u4f4e\u7cfb\u7edf\u7684\u53ef\u7ef4\u62a4\u6027\u548c\u7528\u6237\u7a0b\u5e8f\u7684\u53ef\u4f38\u7f29\u6027\u3002\u56e0\u6b64\uff0c\u4e1a\u754c\u5df2\u7ecf\u5f00\u53d1\u4e86\u8bb8\u591a\u4e13\u95e8\u7684\u5b9e\u65f6\u6d41\u6570\u636e\u5904\u7406\u7cfb\u7edf\u6765\u6ee1\u8db3\u8fd9\u4e9b\u9700\u6c42\u3002 1\u3001\u7f16\u7a0b\u5b9e\u73b0\u6587\u4ef6\u5408\u5e76\u548c\u53bb\u91cd\u64cd\u4f5c \u5bf9\u4e8e\u4e24\u4e2a\u8f93\u5165\u6587\u4ef6\uff0c\u5373\u6587\u4ef6 A \u548c\u6587\u4ef6 B\uff0c\u8bf7\u7f16\u5199 MapReduce \u7a0b\u5e8f\uff0c\u5bf9\u4e24\u4e2a\u6587\u4ef6\u8fdb\u884c\u5408\u5e76\uff0c\u5e76\u5254\u9664\u5176\u4e2d\u91cd\u590d\u7684\u5185\u5bb9\uff0c\u5f97\u5230\u4e00\u4e2a\u65b0\u7684\u8f93\u51fa\u6587\u4ef6 C\u3002\u4e0b\u9762\u662f\u8f93\u5165\u6587\u4ef6\u548c\u8f93\u51fa\u6587\u4ef6\u7684\u4e00\u4e2a\u6837\u4f8b\uff0c\u4f9b\u53c2\u8003\u3002 \u8f93\u5165\u6587\u4ef6A\u7684\u6837\u4f8b\u5982\u4e0b\uff1a 20150101 x 20150102 y 20150103 x 20150104 y 20150105 z 20150106 x \u8f93\u5165\u6587\u4ef6B\u7684\u6837\u4f8b\u5982\u4e0b\uff1a [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[67],"tags":[],"class_list":["post-289","post","type-post","status-publish","format-standard","hentry","category-67"],"_links":{"self":[{"href":"https:\/\/solost.cn\/index.php?rest_route=\/wp\/v2\/posts\/289","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/solost.cn\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/solost.cn\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/solost.cn\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/solost.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=289"}],"version-history":[{"count":0,"href":"https:\/\/solost.cn\/index.php?rest_route=\/wp\/v2\/posts\/289\/revisions"}],"wp:attachment":[{"href":"https:\/\/solost.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=289"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/solost.cn\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=289"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/solost.cn\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=289"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}