Subset-sum via map-reduce
andrew.musselman@gmail.com

To run this using bash and pipes do 

./run.sh

It will call generate-numbers.py, which creates a file of unique random integers, then 
will output any combinations of numbers that sum to zero.

You can adjust the parameters for generation of random numbers by editing generate-numbers.py.



To run this using hadoop-streaming, make sure you have a cluster at least in pseudo-distrubuted 
mode running, then do

./run-hs.sh

To run it again, delete the output directories and files from your HDFS and re-run by doing

$ for dir in `hadoop fs -ls | cut -b 66-`; do hadoop fs -rmr $dir; done; time ./run-hs.sh