原文链接:http://bjyjtdj.iteye.com/blog/1454803
本文讲述如何在map端完成join操作。之前我们提到了reduce-join,这种方法的灵活性不错,也是理所当然地能够想到的方法;但这种方法存在的一个最大的问题是性能。大量的中间数据需要从map节点通过网络发送到reduce节点,因而效率比较低。实际上,两表的join操作中很多都是无用的数据。现在考虑可能的一种场景,其中一个表非常小,以致于可以直接存放在内存中,那么我们可以利用Hadoop提供的DistributedCache机制,将较小的表加入到其中,在每个map节点都能够访问到该表,最终实现在map阶段完成join操作。这里提一下DistributedCache,可以直观上将它看作是一个全局的只读空间,存储一些需要共享的数据;具体可以参看Hadoop相关资料,这里不进行深入讨论。
实现的源码如下,原理非常简单明了:
- import java.io.BufferedReader;
- import java.io.FileReader;
- import java.io.IOException;
- import java.util.Hashtable;
- import org.apache.hadoop.conf.Configuration;
- import org.apache.hadoop.conf.Configured;
- import org.apache.hadoop.filecache.DistributedCache;
- import org.apache.hadoop.fs.Path;
- 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.KeyValueTextInputFormat;
- import org.apache.hadoop.mapred.MapReduceBase;
- import org.apache.hadoop.mapred.Mapper;
- import org.apache.hadoop.mapred.OutputCollector;
- import org.apache.hadoop.mapred.Reporter;
- import org.apache.hadoop.mapred.TextOutputFormat;
- import org.apache.hadoop.util.Tool;
- import org.apache.hadoop.util.ToolRunner;
- @SuppressWarnings("deprecation")
- public class DataJoinDC extends Configured implements Tool{
- private final static String inputa = "hdfs://m100:9000/joinTest/Customers";
- private final static String inputb = "hdfs://m100:9000/joinTest/Orders";
- private final static String output = "hdfs://m100:9000/joinTest/output";
- public static class MapClass extends MapReduceBase
- implements Mapper<Text, Text, Text, Text> {
- private Hashtable<String, String> joinData = new Hashtable<String, String>();
- @Override
- public void configure(JobConf conf) {
- try {
- Path [] cacheFiles = DistributedCache.getLocalCacheFiles(conf);
- if (cacheFiles != null && cacheFiles.length > 0) {
- String line;
- String[] tokens;
- BufferedReader joinReader = new BufferedReader(
- new FileReader(cacheFiles[0].toString()));
- try {
- while ((line = joinReader.readLine()) != null) {
- tokens = line.split(",", 2);
- joinData.put(tokens[0], tokens[1]);
- }
- }finally {
- joinReader.close();
- }}} catch (IOException e) {
- System.err.println("Exception reading DistributedCache: " + e);
- }
- }
- public void map(Text key, Text value,OutputCollector<Text, Text> output,
- Reporter reporter) throws IOException {
- // for(String t: joinData.keySet()){
- // output.collect(new Text(t), new Text(joinData.get(t)));
- // }
- String joinValue = joinData.get(key.toString());
- if (joinValue != null) {
- output.collect(key,new Text(value.toString() + "," + joinValue));
- }
- }
- }
- @Override
- public int run(String[] args) throws Exception {
- Configuration conf = getConf();
- DistributedCache.addCacheFile(new Path(inputa).toUri(), conf);
- JobConf job = new JobConf(conf, DataJoinDC.class);
- Path in = new Path(inputb);
- Path out = new Path(output);
- FileInputFormat.setInputPaths(job, in);
- FileOutputFormat.setOutputPath(job, out);
- job.setJobName("DataJoin with DistributedCache");
- job.setMapperClass(MapClass.class);
- job.setNumReduceTasks(0);
- job.setInputFormat(KeyValueTextInputFormat.class);
- job.setOutputFormat(TextOutputFormat.class);
- job.set("key.value.separator.in.input.line", ",");
- JobClient.runJob(job);
- return 0;
- }
- public static void main(String[] args) throws Exception{
- int res = ToolRunner.run(new Configuration(), new DataJoinDC(), args);
- System.exit(res);
- }
- }
以上参照《Hadoop in Action》 所附代码,我这里是将Customers表作为较小的表,传入DistributedCache。
这里需要注意的地方
- DistributedCache.addCacheFile(new Path(inputa).toUri(), conf);
这句一定要放在job初始化之前,否则在map中读取不到文件。因为job初始化时将传入Configuration对象拷贝了一份给了JobContext!