MapReduce Example

Let’s use the MovieLens dataset as an example and find out how many movies did each user rated.

The MAPPER converts raw source data into key/value pairs

This is how the MovieLens u.data looks like:

• Map users to movies they watched:

• Extract and organize data we care about.
• The less data we put on the cluster, the better.

• MapReduce sorts and groups the mapped data (“Shuffle and Sort”)

• The REDUCER Processes each key’s values

• To summarize:

• Example on a cluster: