The purpose of this presentation is to provide the basic concept of MapReduce programming model by some classic examples from Hadoop and Spark. The other purpose is to show that MapReduce is a foundation for solving big data using modern and powerful Spark API. For solving big data problems, our experience indicates that MapReduce by itself (as a series of map and reduce functions) is not sufficient to address all types of problems. Clearly, Spark API (as a superset of MapReduce paradigm) addresses the big data problem as a general compute engine (by providing basic MapReduce as well as other powerful features such as join(), filter(), cartesian(), and combineByKey()).

 To be provided at a later time.

 Seating is limited. RSVP via email to compsci-gsa@ncsu.edu