通过hadoop上的hive完成WordCount

通过hadoop上的hive完成WordCount

启动hadoop

1
start - all .sh

Hdfs上创建文件夹

1
2
hdfs dfs  - mkdir songinput
hdfs dfs  - ls  / user / hadoop

上传文件至hdfs:

下载歌词songs.txt保存在~/hadoop里,查询目录

1
2
cd ~ / hadoop
ls

  

上传至hdfs

1
2
hdfs dfs  - put . / songs.txt songinput
hdfs dfs  - ls  / user / hadoop / songinput

 

启动Hive

1
hive

创建原始文档表

1
create table docs(line string);<br>show tables;

  

导入文件内容到表docs并查看

1
load data inpath  '/user/hadoop/songinput/songs.txt'  overwrite into table docs;<br>select  *  from  docs;

用HQL进行词频统计,结果放在表word_count里

1
create table word_count as select word,count( 1 from  (select explode(split(line, ' ' )) as word  from  docs) word group by word;

查看统计结果

1
select  *  from  word_count;

猜你喜欢

转载自www.cnblogs.com/leonHQ/p/9047717.html