就用单词计数这个例子,需要统计的单词存在HBase中的word表,MapReduce执行的时候从word表读取数据,统计结束后将结果写入到HBase的stat表中。
1、在eclipse中建立一个hadoop项目,然后从hbase的发布包中引入如下jar
hbase-0.94.13.jar zookeeper-3.4.5.jar protobuf-java-2.4.0a.jar guava-11.0.2.jar
2、在HBase中建立相关的表和初始化测试数据
package cn.luxh.app; import java.io.IOException; import java.util.ArrayList; import java.util.List; import org.apache.hadoop.hbase.client.HTable; import org.apache.hadoop.hbase.client.Put; /** * * @author Luxh * */ public class InitData { public static void main(String[] args) throws IOException { //创建一个word表,只有一个列族content HBaseUtil.createTable("word","content"); //获取word表 HTable htable = HBaseUtil.getHTable("word"); htable.setAutoFlush(false); //创建测试数据 List<Put> puts = new ArrayList<Put>(); Put put1 = HBaseUtil.getPut("1","content",null,"The Apache Hadoop software library is a framework"); Put put2 = HBaseUtil.getPut("2","content",null,"The common utilities that support the other Hadoop modules"); Put put3 = HBaseUtil.getPut("3","content",null,"Hadoop by reading the documentation"); Put put4 = HBaseUtil.getPut("4","content",null,"Hadoop from the release page"); Put put5 = HBaseUtil.getPut("5","content",null,"Hadoop on the mailing list"); puts.add(put1); puts.add(put2); puts.add(put3); puts.add(put4); puts.add(put5); //提交测试数据 htable.put(puts); htable.flushCommits(); htable.close(); //创建stat表,只有一个列祖result HBaseUtil.createTable("stat","result"); } }
1)代码中的HBaseUtil工具类参考:http://www.cnblogs.com/luxh/archive/2013/04/16/3025172.html
2)执行上面的程序后,查看HBase中是否已创建成功
hbase(main):012:0> list
TABLE
stat
word
2 row(s) in 0.4730 seconds
3)查看word中的测试数据
hbase(main):005:0> scan 'word' ROW COLUMN+CELL 1 column=content:, timestamp=1385447676510, value=The Apache Hadoo p software library is a framework 2 column=content:, timestamp=1385447676510, value=The common utili ties that support the other Hadoop modules 3 column=content:, timestamp=1385447676510, value=Hadoop by readin g the documentation 4 column=content:, timestamp=1385447676510, value=Hadoop from the release page 5 column=content:, timestamp=1385447676510, value=Hadoop on the ma iling list 5 row(s) in 5.7810 seconds
3、MapReduce程序
package cn.luxh.app; import java.io.IOException; import java.util.StringTokenizer; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.hbase.HBaseConfiguration; import org.apache.hadoop.hbase.client.Put; import org.apache.hadoop.hbase.client.Result; import org.apache.hadoop.hbase.client.Scan; import org.apache.hadoop.hbase.io.ImmutableBytesWritable; import org.apache.hadoop.hbase.mapreduce.TableMapReduceUtil; import org.apache.hadoop.hbase.mapreduce.TableMapper; import org.apache.hadoop.hbase.mapreduce.TableReducer; import org.apache.hadoop.hbase.util.Bytes; import org.apache.hadoop.io.IntWritable; import org.apache.hadoop.io.Text; import org.apache.hadoop.mapreduce.Job; /** * @author Luxh * */ public class WordStat { /** * TableMapper<Text,IntWritable> Text:输出的key类型,IntWritable:输出的value类型 */ public static class MyMapper extends TableMapper<Text,IntWritable>{ private static IntWritable one = new IntWritable(1); private static Text word = new Text(); @Override protected void map(ImmutableBytesWritable key, Result value, Context context) throws IOException, InterruptedException { //表里面只有一个列族,所以我就直接获取每一行的值 String words = Bytes.toString(value.list().get(0).getValue()); StringTokenizer st = new StringTokenizer(words); while (st.hasMoreTokens()) { String s = st.nextToken(); word.set(s); context.write(word, one); } } } /** * TableReducer<Text,IntWritable> Text:输入的key类型,IntWritable:输入的value类型,ImmutableBytesWritable:输出类型 */ public static class MyReducer extends TableReducer<Text,IntWritable,ImmutableBytesWritable>{ @Override protected void reduce(Text key, Iterable<IntWritable> values, Context context) throws IOException, InterruptedException { int sum = 0; for(IntWritable val:values) { sum+=val.get(); } //添加一行记录,每一个单词作为行键 Put put = new Put(Bytes.toBytes(key.toString())); //在列族result中添加一个标识符num,赋值为每个单词出现的次数 //String.valueOf(sum)先将数字转化为字符串,否则存到数据库后会变成\x00\x00\x00\x这种形式 //然后再转二进制存到hbase。 put.add(Bytes.toBytes("result"), Bytes.toBytes("num"), Bytes.toBytes(String.valueOf(sum))); context.write(new ImmutableBytesWritable(Bytes.toBytes(key.toString())),put); } } public static void main(String[] args) throws IOException, ClassNotFoundException, InterruptedException { Configuration conf = HBaseConfiguration.create(); Job job = new Job(conf,"wordstat"); job.setJarByClass(Blog.class); Scan scan = new Scan(); //指定要查询的列族 scan.addColumn(Bytes.toBytes("content"),null); //指定Mapper读取的表为word TableMapReduceUtil.initTableMapperJob("word", scan, MyMapper.class, Text.class, IntWritable.class, job);
//指定Reducer写入的表为stat TableMapReduceUtil.initTableReducerJob("stat", MyReducer.class, job); System.exit(job.waitForCompletion(true)?0:1); } }
等待程序执行结束,查看统计表stat
hbase(main):014:0> scan 'stat' ROW COLUMN+CELL Apache column=result:num, timestamp=1385449492309, value=1 Hadoop column=result:num, timestamp=1385449492309, value=5 The column=result:num, timestamp=1385449492309, value=2 a column=result:num, timestamp=1385449492309, value=1 by column=result:num, timestamp=1385449492309, value=1 common column=result:num, timestamp=1385449492309, value=1 documentation column=result:num, timestamp=1385449492309, value=1 framework column=result:num, timestamp=1385449492309, value=1 from column=result:num, timestamp=1385449492309, value=1 is column=result:num, timestamp=1385449492309, value=1 library column=result:num, timestamp=1385449492309, value=1 list column=result:num, timestamp=1385449492309, value=1 mailing column=result:num, timestamp=1385449492309, value=1 modules column=result:num, timestamp=1385449492309, value=1 on column=result:num, timestamp=1385449492309, value=1 other column=result:num, timestamp=1385449492309, value=1 page column=result:num, timestamp=1385449492309, value=1 reading column=result:num, timestamp=1385449492309, value=1 release column=result:num, timestamp=1385449492309, value=1 software column=result:num, timestamp=1385449492309, value=1 support column=result:num, timestamp=1385449492309, value=1 that column=result:num, timestamp=1385449492309, value=1 the column=result:num, timestamp=1385449492309, value=4 utilities column=result:num, timestamp=1385449492309, value=1 24 row(s) in 0.7970 seconds