The three java files are written in map-reduce.

Function:

wordcount.java

This file can be found in any hadoop source, just the "hello word" for hadoop. The function of this is to count one word frequency in a given file.

wcdouble.java

This java file aims at counting double-word in a given file.

ucache.java

This java file can be used to count certain parrent words in given files. There fore, I used distributed cache to store the pattern file and then to count the word exists in the file.

use statement below to utilize ucache.jar

bin/hadoop jar ucache.jar ucache ¨Cslist ~/patterns.txt ~/input Output