使用dumbo开发hadoop streaming程序

1. dumbo的官网：https://github.com/klbostee/dumbo/

wiki: https://github.com/klbostee/dumbo/wiki

2. 安装配置

详见wiki:https://github.com/klbostee/dumbo/wiki/Building-and-installing

解压从网站下载下来的源码到dumbo文件夹，执行如下命令就可以了

1	wget -O ez_setup.py  http://peak.telecommunity.com/dist/ez_setup.py

2	python ez_setup.py dumbo

3. 例子程序

使用dumbo开发程序来统计apache的访问日志中最多的几个ip地址

apache日志格式如下：

1	[admin@search011 dumbotest]$ head -n3 access.log

2	10.0.0.1 - - [02/Nov/2010:10:15:52 +0800] "GET /index/Back/task_updator.php?taskid=22870&sub_taskid=15&runhost=localhost&status=OK HTTP/1.0" 200 3 "-" "Wget/1.10.2 (Red Hat modified)"

3	10.0.0.3 - - [02/Nov/2010:10:15:52 +0800] "GET /index/Back/task_updator.php?taskid=22870&sub_taskid=17&runhost=localhost&status=OK HTTP/1.0" 200 3 "-" "Wget/1.10.2 (Red Hat modified)"

4	10.0.0.1 - - [02/Nov/2010:10:15:53 +0800] "GET /index/Back/task_updator.php?taskid=22870&sub_taskid=9&runhost=localhost&status=OK HTTP/1.0" 200 3 "-" "Wget/1.10.2 (Red Hat modified)"

ipcount.py程序如下：

1	#!/bin/env python

2

3	def mapper(key, value):

4	yield value.split(" ")[0], 1

5

6	def reducer(key, values):

7	yield key, sum(values)

8

9	if __name__ == "__main__":

10	import dumbo

11	dumbo.run(mapper, reducer, combiner=reducer)

运行：

1	cut -d ' ' -f 1 access.log | sort | uniq -c | sort -nr | head -n 5

可以看到在本机上执行的结果。

在hadoop上运行：

1	dumbo start ipcount.py -hadoop /home/admin/hadoop -libjar /home/admin/hadoop/contrib/streaming/hadoop-streaming-0.20.2-CDH3B4.jar  -input /admin/input -output /admin/output

1	就可以提交job看到运行结果了。