多路归并排序【JAVA实现】

对远远大于内存的数据进行外排序，在多路比较的时候用败者树效率会更高。
package my.sort;
import java.io.BufferedInputStream;
import java.io.BufferedOutputStream;
import java.io.BufferedWriter;
import java.io.DataInputStream;
import java.io.DataOutputStream;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileNotFoundException;
import java.io.FileOutputStream;
import java.io.FileWriter;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Iterator;
import java.util.Random;
/**
* 基于大数据量的外排序算法，分为二路贵宾和多路归并
* @author java2king
* @link http://blog.csdn.net/Java2King
*
*/
public class ExternalSort {
public static int ITEM_COUNT = 10000000; //总数
public static int BUFFER_SIZE = 1024*4*1000;// 一次缓冲读取

public static int FILE_COUNT = 1024*1000*1*4;// 每个文件的记录数1

public static File MAIN_FILE = new File("mainset");//要排序的文件
/**
* 二路归并
* @param file
* @return
* @throws IOException
*/
public File sort(File file) throws IOException {
ArrayList<File> files = split(file);
return process(files);
}
/**
* 多路归并
* @param file
* @throws IOException
*/
public void mSort(File file) throws IOException{
ArrayList<File> files = split(file);
multipleMerge(files);

}
// recursive method to merge the lists until we are left with a
// single merged list
private File process(ArrayList<File> list) throws IOException {
if (list.size() == 1) {
return list.get(0);
}
ArrayList<File> inter = new ArrayList<File>();
for (Iterator<File> itr = list.iterator(); itr.hasNext();) {
File one = itr.next();
if (itr.hasNext()) {
File two = itr.next();
inter.add(merge(one, two)