Some examples and notes about writing map reduces in ruby to run under Hadoop.
- Counting Words (got from Big Fast Blog)
- Running locally without Apache Hadoop
You can test your maps/reduces on your shell prompt by just piping the source data from the source file to the map, sorting its output before piping it to the reduce.
cat my-source-file.txt | ruby map.rb | sort | ruby reduce.rb
- The output of a map
Each output of a map must always be on the format key tab value newline
puts "#{key}\t#{value}"