理解MapReduce计算构架

用Python编写WordCount程序任务

程序	WordCount
输入	一个包含大量单词的文本文件
输出	文件中每个单词及其出现次数（频数），并按照单词字母顺序排序，每个单词和其频数占一行，单词和频数之间有间隔

1. 编写map函数，reduce函数
2. 将其权限作出相应修改
3. 本机上测试运行代码
4. 放到HDFS上运行
5. 下载并上传文件到hdfs上
6. 用Hadoop Streaming命令提交任务

   1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

   create 'Student', ' S_No  ','S_Name', 'S_Sex','S_Age'   put 'Student','s001','S_No','2015001' put 'Student','s001','S_Name','Zhangsan' put 'Student','s001','S_Sex','male' put 'Student','s001','S_Age','23'   put 'Student','s002','S_No','2015003' put 'Student','s002','S_Name','Mary' put 'Student','s002','S_Sex','female' put 'Student','s002','S_Age','22'   put 'Student','s003','S_No','2015003' put 'Student','s003','S_Name','Lisi' put 'Student','s003','S_Sex','male' put 'Student','s003','S_Age','24'

   　　

   1 2 3 4 5 6

   scan 'Student' alter 'Student','NAME'=>'course' put 'Student','3','course:Math','85' dorp 'Student','course' count 's1' truncate 's1'

   　　

   1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37

   cd /home/hadoop/wc sudo gedit mapper.py   # map函数 import sys for i in stdin:     i = i.strip()     words = i.split()     for word in words:     print '%s %s' % (word,1)   #reduce函数 from operator import itemgetter import sys   current_word = None current_count = 0 word = None   for i in stdin:     i = i.strip()     word, count = i.split(' ',1)     try:     count = int(count)     except ValueError:     continue       if current_word == word:     current_count += count     else:     if current_word:         print '%s %s' % (current_word, current_count)     current_count = count     current_word = word   if current_word == word:     print '%s %s' % (current_word, current_count)

   　　

   1	chmod a+x /home/hadoop/mapper.py

   　　

   1 2 3

   echo "foo foo quux labs foo bar quux" | /home/hadoop/wc/mapper.py   echo "foo foo quux labs foo bar quux" | /home/hadoop/wc/mapper.py | sort -k1,1 | /home/hadoop/wc/reducer.p

   　　

   1 2 3 4 5 6 7

   cd  /home/hadoop/wc wget http://www.gutenberg.org/files/5000/5000-8.txt wget http://www.gutenberg.org/cache/epub/20417/pg20417.txt     cd /usr/hadoop/wc hdfs dfs -put /home/hadoop/hadoop/gutenberg/*.txt /user/hadoop/input