hadoop之mapreduce实例

hadoop之mapreduce实例

例子很简单，我没有运行自带的wordcount，而是自己做了一个简单的例子。
实现的功能是从我们的nginx的access log里面计算url访问的次数。
access log文件：

10.2.112.22 - -  [ 11/Apr/2012:10:25:31 +0800 ] "GET /bf5bd91c/css/base/base_jiexi-all-min.css HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:31 +0800 ] "GET /bf5bd91c/js/lib/lib-min.js HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:31 +0800 ] "GET /image/jiexi/logo.png HTTP/1.1" 304 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:31 +0800 ] "GET /bf5bd91c/js/page/jiexi/index-min.js HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:31 +0800 ] "GET /bf5bd91c/css/page/jiexi/index-all-min.css HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/css/page/jiexi/index-all-min.css HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/js/page/jiexi/index-min.js HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/js/lib/lib-min.js HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/css/base/base_jiexi-all-min.css HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/css/page/jiexi/index-all-min.css HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/js/page/jiexi/index-min.js HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/js/page/jiexi/index-min.js HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/css/page/jiexi/index-all-min.css HTTP/1.1" 499 0 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/css/base/base_jiexi-all-min.css HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/js/lib/lib-min.js HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/css/page/jiexi/index-all-min.css HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /bf5bd91c/js/page/jiexi/index-min.js HTTP/1.1" 302 161 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/js/page/jiexi/index-min.js HTTP/1.1" 200 56215 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/css/page/jiexi/index-all-min.css HTTP/1.1" 200 21254 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/css/base/base_jiexi-all-min.css HTTP/1.1" 200 22782 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"
10.2.112.22 - -  [ 11/Apr/2012:10:25:32 +0800 ] "GET /release/js/lib/lib-min.js HTTP/1.1" 200 137514 "http://www.jiexi.com/home" "Mozilla/5.0 (Windows; U; Windows NT 6.1; zh-CN; rv:1.9.2.28) Gecko/20120306 Firefox/3.6.28"

新建maven项目：

< project  xmlns ="http://maven.apache.org/POM/4.0.0"  xmlns:xsi ="http://www.w3.org/2001/XMLSchema-instance"
    xsi:schemaLocation ="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd" >
     < modelVersion >4.0.0 </ modelVersion >
     < groupId >com.jiexi </ groupId >
     < artifactId >jiexi-examples </ artifactId >
     < version >0.0.1-SNAPSHOT </ version >
     < dependencies >
         < dependency >
             < groupId >org.apache.hadoop </ groupId >
             < artifactId >hadoop-core </ artifactId >
             < version >1.0.2 </ version >
         </ dependency >
     </ dependencies >
</ project >

Mapper代码如下：

package com.jiexi.examples.hadoop;

import java.io.IOException;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.LongWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.MapReduceBase;
import org.apache.hadoop.mapred.Mapper;
import org.apache.hadoop.mapred.OutputCollector;
import org.apache.hadoop.mapred.Reporter;

public class AccessLogMapper extends MapReduceBase implements
        Mapper < LongWritable , Text, Text, IntWritable > {
    private final static IntWritable one = new IntWritable(1);
    private Text url = new Text();

    static String POST = "\"POST ";
    static String GET = "\"GET ";
    static String END = " HTTP/1.0";

    public void map(LongWritable key, Text value,
            OutputCollector < Text , IntWritable > output, Reporter reporter)
            throws IOException {
        String line = value.toString();
        String url1 = getUrl(line);

        url.set(url1);
        output.collect(url, one);
    }

    public static void main(String[] args) {
        String a = "10.2.112.34 - - [06/Mar/2012:18:05:41 +0800] \"GET /mine?originUrl= HTTP/1.0\" 302 -";
        String b = "10.2.112.34 - - [06/Mar/2012:15:02:42 +0800] \"POST /user/login?originUrl=http%3A%2F%2Fwww.jiexi.com%2Fhome HTTP/1.0\" 200 25";
 
//        System.out.println(getUrl(a));
//        System.out.println(getUrl(b));
        
        String s =" /user/register?originUrl=http%3A%2F%2Fwww.jiexi.com%2Fhome";
        
        System.out.println(s.substring(0,s.indexOf("?")));
    }

    private static String getUrl(String a) {
        // int len = POST.length();
        int begin = a.indexOf(POST);
        int get = a.indexOf(GET);
        if (get > -1) {
            begin = get;
            // len = GET.length();
        }

        int end = a.indexOf(END);

        String url = a.substring(begin + 1, end);

        if (url.indexOf("?") > 0) {
            return url.substring(0, url.indexOf("?"));
        }

        return url;
    }

}

Reducer代码如下：

package com.jiexi.examples.hadoop;

import java.io.IOException;
import java.util.Iterator;

import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.MapReduceBase;
import org.apache.hadoop.mapred.OutputCollector;
import org.apache.hadoop.mapred.Reducer;
import org.apache.hadoop.mapred.Reporter;

public  class AccessLogReducer  extends MapReduceBase  implements
        Reducer<Text, IntWritable, Text, IntWritable> {

     public  void reduce(Text key, Iterator<IntWritable> values,
            OutputCollector<Text, IntWritable> output, Reporter reporter)
             throws IOException {
         int sum = 0;
         while (values.hasNext()) {
            sum += values.next().get();
        }
        output.collect(key,  new IntWritable(sum));
    }

}

job调用

package com.jiexi.examples.hadoop;

import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IntWritable;
import org.apache.hadoop.io.Text;
import org.apache.hadoop.mapred.FileInputFormat;
import org.apache.hadoop.mapred.FileOutputFormat;
import org.apache.hadoop.mapred.JobClient;
import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.mapred.TextInputFormat;
import org.apache.hadoop.mapred.TextOutputFormat;

public  class AccessLogPerDayJob {
     public  static  void main(String[] args)  throws Throwable {
        JobConf jobConf =  new JobConf(AccessLogPerDayJob. class);
        jobConf.setJobName("access_log");

        jobConf.setOutputKeyClass(Text. class);
        jobConf.setOutputValueClass(IntWritable. class);

        jobConf.setMapperClass(AccessLogMapper. class);
        jobConf.setCombinerClass(AccessLogReducer. class);
        jobConf.setReducerClass(AccessLogReducer. class);

        jobConf.setInputFormat(TextInputFormat. class);
        jobConf.setOutputFormat(TextOutputFormat. class);

        FileInputFormat.addInputPath(jobConf,  new Path(args[0]));
        FileOutputFormat.setOutputPath(jobConf,  new Path(args[1]));

        JobClient.runJob(jobConf);
    }
}

===============================================
mvn clean package 
丢到namenode的/opt/hadoop 下面

运行：
./bin/hadoop dfs -mkdir access_log_in
./bin/hadoop dfs -ls

#拷贝本地日志文件到hdfs中
./bin/hadoop dfs -put /opt/access_log/*  access_log_in

#运行，把access_log.jar拷贝到/opt/hadoop下面
./bin/hadoop jar access_log.jar com.jiexi.examples.hadoop.AccessLogPerDayJob    access_log_in  access_log_out


查看job运行情况：
http://10.2.112.31:50030/jobtracker.jsp 

查看data：
./bin/hadoop dfs -ls access_log_out