Steps are as follows: 

1 cp all nutch conf(only one single nutch instance should be setup) to hadoop conf

 2 put urls and /nutch-1.3.job(jar) to cloud 
 3 use the crawl command with reference files on the cloud