众所周知,Hadoop对处理单个大文件比处理多个小文件更有效率,另外单个文件也非常占用HDFS的存储空间。所以往往要将其合并起来。

1,getmerge

hadoop有一个命令行工具getmerge,用于将一组HDFS上的文件复制到本地计算机以前进行合并

参考:http://hadoop.apache.org/common/docs/r0.19.2/cn/hdfs_shell.html

使用方法:hadoop fs -getmerge <src> <localdst> [addnl]

接受一个源目录和一个目标文件作为输入,并且将源目录中所有的文件连接成本地目标文件。addnl是可选的,用于指定在每个文件结尾添加一个换行符。

多嘴几句:调用文件系统(FS)Shell命令应使用 bin/hadoop fs <args>的形式。 所有的的FS shell命令使用URI路径作为参数。URI格式是scheme://authority/path

2.putmerge

将本地小文件合并上传到HDFS文件系统中。

一种方法可以现在本地写一个脚本,先将一个文件合并为一个大文件,然后将整个大文件上传,这种方法占用大量的本地磁盘空间;

另一种方法如下,在复制的过程中上传。参考:《hadoop in action》

import java.io.IOException;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataInputStream;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileStatus;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IOUtils;

//参数1为本地目录,参数2为HDFS上的文件
public class PutMerge {
	
	public static void putMergeFunc(String LocalDir, String fsFile) throws IOException
	{
		Configuration  conf = new Configuration();
		FileSystem fs = FileSystem.get(conf);		//fs是HDFS文件系统
		FileSystem local = FileSystem.getLocal(conf);	//本地文件系统
		
		Path localDir = new Path(LocalDir);
		Path HDFSFile = new Path(fsFile);
		
		FileStatus[] status =  local.listStatus(localDir);	//得到输入目录
		FSDataOutputStream out = fs.create(HDFSFile);		//在HDFS上创建输出文件
		
		for(FileStatus st: status)
		{
			Path temp = st.getPath();
			FSDataInputStream in = local.open(temp);
			IOUtils.copyBytes(in, out, 4096, false);	//读取in流中的内容放入out
			in.close();	//完成后,关闭当前文件输入流
		}
		out.close();
	}
	public static void main(String [] args) throws IOException
	{
		String l = "/home/kqiao/hadoop/MyHadoopCodes/putmergeFiles";
		String f = "hdfs://ubuntu:9000/user/kqiao/test/PutMergeTest";
		putMergeFunc(l,f);
	}
}

3.将小文件打包成SequenceFile的MapReduce任务

来自:《hadoop权威指南》

实现将整个文件作为一条记录处理的InputFormat:

public class WholeFileInputFormat
    extends FileInputFormat<NullWritable, BytesWritable> {
  
  @Override
  protected boolean isSplitable(JobContext context, Path file) {
    return false;
  }

  @Override
  public RecordReader<NullWritable, BytesWritable> createRecordReader(
      InputSplit split, TaskAttemptContext context) throws IOException,
      InterruptedException {
    WholeFileRecordReader reader = new WholeFileRecordReader();
    reader.initialize(split, context);
    return reader;
  }
}

实现上面类中使用的定制的RecordReader:

/实现一个定制的RecordReader,这六个方法均为继承的RecordReader要求的虚函数。
//实现的RecordReader,为自定义的InputFormat服务
public class WholeFileRecordReader extends RecordReader<NullWritable, BytesWritable>{

	private FileSplit fileSplit;
	private Configuration conf;
	private BytesWritable value = new BytesWritable();
	private boolean processed = false;
	@Override
	public void close() throws IOException {
		// do nothing
	}

	@Override
	public NullWritable getCurrentKey() throws IOException,
			InterruptedException {
		return NullWritable.get();
	}

	@Override
	public BytesWritable getCurrentValue() throws IOException,
			InterruptedException {
		return value;
	}

	@Override
	public float getProgress() throws IOException, InterruptedException {
		return processed? 1.0f : 0.0f;
	}

	@Override
	public void initialize(InputSplit split, TaskAttemptContext context)
			throws IOException, InterruptedException {
		this.fileSplit = (FileSplit) split;
		this.conf = context.getConfiguration();
	}

	//process表示记录是否已经被处理过
	@Override
	public boolean nextKeyValue() throws IOException, InterruptedException {
	    if (!processed) {
	        byte[] contents = new byte[(int) fileSplit.getLength()];
	        Path file = fileSplit.getPath();
	        FileSystem fs = file.getFileSystem(conf);
	        FSDataInputStream in = null;
	        try {
	          in = fs.open(file);
	          	              //将file文件中 的内容放入contents数组中。使用了IOUtils实用类的readFully方法,将in流中得内容放入
	          //contents字节数组中。
	          IOUtils.readFully(in, contents, 0, contents.length);
	          //BytesWritable是一个可用做key或value的字节序列,而ByteWritable是单个字节。
	          					//将value的内容设置为contents的值
	          value.set(contents, 0, contents.length);
	        } finally {
	          IOUtils.closeStream(in);
	        }
	        processed = true;
	        return true;
	      }
	      return false;
	}
}

将小文件打包成SequenceFile:

public class SmallFilesToSequenceFileConverter extends Configured implements Tool{

	//静态内部类,作为mapper
	static class SequenceFileMapper extends Mapper<NullWritable, BytesWritable, Text, BytesWritable>
	{
		private Text filenameKey;
		
		//setup在task开始前调用,这里主要是初始化filenamekey
		@Override
		protected void setup(Context context)
		{
			InputSplit split = context.getInputSplit();
			Path path = ((FileSplit) split).getPath();
			filenameKey = new Text(path.toString());
		}
		@Override
		public void map(NullWritable key, BytesWritable value, Context context)
				throws IOException, InterruptedException{
			context.write(filenameKey, value);
		}
	}

	@Override
	public int run(String[] args) throws Exception {
		Configuration conf = new Configuration();
		Job job = new Job(conf);
		job.setJobName("SmallFilesToSequenceFileConverter");
		
		FileInputFormat.addInputPath(job, new Path(args[0]));
		FileOutputFormat.setOutputPath(job, new Path(args[1]));
		
		//再次理解此处设置的输入输出格式。。。它表示的是一种对文件划分,索引的方法
		job.setInputFormatClass(WholeFileInputFormat.class);
		job.setOutputFormatClass(SequenceFileOutputFormat.class);
		
		//此处的设置是最终输出的key/value,一定要注意!
		job.setOutputKeyClass(Text.class);
		job.setOutputValueClass(BytesWritable.class);
		
		job.setMapperClass(SequenceFileMapper.class);
		
		return job.waitForCompletion(true) ? 0 : 1;
	}
	
	public static void main(String [] args) throws Exception
	{
		int exitCode = ToolRunner.run(new SmallFilesToSequenceFileConverter(), args);
		System.exit(exitCode);
	}
}