Hadoop流 虽然Hadoop是用java写的,但是Hadoop提供了Hadoop流,Hadoop流提供一个API, 允许用户使用任何语言编写map函数和reduce函数. Hadoop流动关键是,它使用UNIX标准流作为程序与Hadoop之间的接口。因此,任何程序只要可以从标准输入流中读取数据,并且可以把数据写入标准输出流中,那么就可以通过Hadoop流使用任何语言编写MapReduce程序的map函数和reduce函数。 例如:bin/hadoop jar contrib/streaming/hadoop-streaming-0.20.203.0.jar -mapper /usr/local/hadoop/mapper.php -reducer /usr/local/hadoop/reducer.php -input test/* -output out4 Hadoop流引入的包:hadoop-streaming-0.20.203.0.jar,Hadoop根目录下是没有hadoop-streaming.jar的,因为streaming是一个contrib,所以要去contrib下面找,以hadoop-0.20.2为例,它在这里: -input:指明输入hdfs文件的路径 -output:指明输出hdfs文件的路径 -mapper:指明map函数 -reducer:指明reduce函数 mapper函数 mapper.php文件,写入如下代码: [php] view plaincopyprint? 01.#!/usr/local/php/bin/php 02.<?php 03.$word2count = array(); 04.// input comes from STDIN (standard input) 05.// You can this code :$stdin = fopen(“php://stdin”, “r”); 06.while (($line = fgets(STDIN)) !== false) { 07. // remove leading and trailing whitespace and lowercase 08. $line = strtolower(trim($line)); 09. // split the line into words while removing any empty string 10. $words = preg_split('/\W/', $line, 0, PREG_SPLIT_NO_EMPTY); 11. // increase counters 12. foreach ($words as $word) { 13. $word2count[$word] += 1; 14. } 15.} 16.// write the results to STDOUT (standard output) 17.// what we output here will be the input for the 18.// Reduce step, i.e. the input for reducer.py 19.foreach ($word2count as $word => $count) { 20. // tab-delimited 21. echo $word, chr(9), $count, PHP_EOL; 22.} 23.?> #!/usr/local/php/bin/php <?php $word2count = array(); // input comes from STDIN (standard input) // You can this code :$stdin = fopen(“php://stdin”, “r”); while (($line = fgets(STDIN)) !== false) { // remove leading and trailing whitespace and lowercase $line = strtolower(trim($line)); // split the line into words while removing any empty string $words = preg_split('/\W/', $line, 0, PREG_SPLIT_NO_EMPTY); // increase counters foreach ($words as $word) { $word2count[$word] += 1; } } // write the results to STDOUT (standard output) // what we output here will be the input for the // Reduce step, i.e. the input for reducer.py foreach ($word2count as $word => $count) { // tab-delimited echo $word, chr(9), $count, PHP_EOL; } ?> 这段代码的大致意思是:把输入的每行文本中的单词找出来,并以” hello 1 world 1″ 这样的形式输出出来。 和之前写的PHP基本没有什么不同,对吧,可能稍微让你感到陌生有两个地方: PHP作为可执行程序 第一行的 [php] view plaincopyprint? 01.#!/usr/local/php/bin/php #!/usr/local/php/bin/php告诉linux,要用#!/usr/local/php/bin/php这个程序作为以下代码的解释器。写过linux shell的人应该很熟悉这种写法了,每个shell脚本的第一行都是这样: #!/bin/bash, #!/usr/bin/python 有了这一行,保存好这个文件以后,就可以像这样直接把mapper.php当作cat, grep一样的命令执行了:./mapper.php 使用stdin接收输入 PHP支持多种参数传入的方法,大家最熟悉的应该是从$_GET, $_POST超全局变量里面取通过Web传递的参数,次之是从$_SERVER['argv']里取通过命令行传入的参数,这里,采用的是标准输入stdin 它的使用效果是: 在linux控制台输入 ./mapper.php mapper.php运行,控制台进入等候用户键盘输入状态 用户通过键盘输入文本 用户按下Ctrl + D终止输入,mapper.php开始执行真正的业务逻辑,并将执行结果输出 那么stdout在哪呢?print本身已经就是stdout啦,跟我们以前写web程序和CLI脚本没有任何不同。 reducer函数 创建reducer.php文件,写入如下代码: [php] view plaincopyprint? 01.#!/usr/local/php/bin/php 02.<?php 03.$word2count = array(); 04.// input comes from STDIN 05.while (($line = fgets(STDIN)) !== false) { 06. // remove leading and trailing whitespace 07. $line = trim($line); 08. // parse the input we got from mapper.php 09. list($word, $count) = explode(chr(9), $line); 10. // convert count (currently a string) to int 11. $count = intval($count); 12. // sum counts 13. if ($count > 0) $word2count[$word] += $count; 14.} 15.// sort the words lexigraphically 16.// 17.// this set is NOT required, we just do it so that our 18.// final output will look more like the official Hadoop 19.// word count examples 20.ksort($word2count); 21.// write the results to STDOUT (standard output) 22.foreach ($word2count as $word => $count) { 23. echo $word, chr(9), $count, PHP_EOL; 24.} 25.?> #!/usr/local/php/bin/php <?php $word2count = array(); // input comes from STDIN while (($line = fgets(STDIN)) !== false) { // remove leading and trailing whitespace $line = trim($line); // parse the input we got from mapper.php list($word, $count) = explode(chr(9), $line); // convert count (currently a string) to int $count = intval($count); // sum counts if ($count > 0) $word2count[$word] += $count; } // sort the words lexigraphically // // this set is NOT required, we just do it so that our // final output will look more like the official Hadoop // word count examples ksort($word2count); // write the results to STDOUT (standard output) foreach ($word2count as $word => $count) { echo $word, chr(9), $count, PHP_EOL; } ?>这段代码的大意是统计每个单词出现了多少次数,并以” hello 2 world 1″ 这样的形式输出 用Hadoop来运行 把文件放入 Hadoop 的 DFS 中: bin/hadoop dfs -put test.log test执行 php 程序处理这些文本(以Streaming方式执行PHP mapreduce程序:): bin/hadoop jar contrib/streaming/hadoop-streaming-0.20.203.0.jar -mapper /usr/local/hadoop/mapper.php -reducer /usr/local/hadoop/reducer.php -input test/* -output out 注意: 1) input和output目录是在hdfs上的路径 2) mapper和reducer是在本地机器的路径,一定要写绝对路径,不要写相对路径,以免到时候hadoop报错说找不到mapreduce程序 3 ) mapper.php 和 reducer.php 必须复制到所有 DataNode 服务器上的相同路径下, 所有的服务器都已经安装php.且安装路径一样. 查看结果 bin/hadoop d fs -cat /tmp/out/part-00000 (责任编辑:IT) |