Post date: Nov 19, 2014 10:25:36 PM
The MapReduce programming paradigm, where the key-value data structure plays a key role, is a surprising simple approach to the big data problem [1, 2]. It includes the following methods.
Map: data -> key-[values]
Reduced: key-[values] -> key-value
[1] http://www.mathworks.com/help/matlab/import_export/getting-started-with-mapreduce.html
[2] http://www.eurecom.fr/~michiard/teaching/slides/clouds/tutorial-mapreduce.pdf