Hi all, I am trying to run wordcount-nopipes.cc. I am using libhdfs to let Hadoop deal with spiltting of files. While debugging I observed that the I am getting the same input for different map tasks.
What do the file names after input split typically look like? Suppose I put a file say 'helloWorld.txt" in HDFS, then what will be the names of the spilts generated be mostly like? Any other ideas, pointers are highly appreciated. Thanks, Prakhar
