Before MapReduce, relational operations like filter, join, sorting, and grouping were the primary operations used for processing large datasets. MapReduce can very easily support operations like filter and sorting. For more information, refer to 2.3.3 Relational-Algebra Operations of the free available book Mining of Massive Datasets, Anand Rajaraman and Jeffrey D. Ullman, Cambridge University Press, 2011.
This recipe explains how to use MapReduce to join two datasets. It will join 100 customers who have bought most items against the dataset that provides items bought by each customer and produce a list of items brought by the 100 most-frequent customers as output.