MongoDB implements the MapReduce framework with its mapReduce() command. An example is shown in Figure 11-6.
Figure 11-6. Running MapReduce in MongoDB
The first two statements define the JavaScript functions map1() and reduce1(). The third statement runs MapReduce on our library.books collection (see Chapter 10, NoSQL Databases), applying those two functions, and naming the resulting collection "map_reduce_example". Appending the find() command causes the output to be displayed.
The map1() function emits the key-value pair (p, 1), where p is the books.publisher field. So this will generate 19 pairs, one for each books document. For example, one of them will be ("OXF", 1). In fact, four of them will be ("MHE", 1), because there are four documents in the books collection whose publisher field is "MHE".
The reduce1() function uses the Array.sum() method to return the sum of the values of the second argument (numBooks) for each value of the first argument (pupId). For example...