Grokking the Aggregation Pipeline
January 24, 2018
Often when developers and DBAs approach MongoDB, they are taught that the Aggregation Pipeline is “How you do SQL GROUP_BY statements in MongoDB”, but then are immediately confused because the construction of an Aggregation Pipeline does not seem to match how they were taught to think through solving problems with SQL.
The first step to mastering a tool is to understand how it is intended to be used, and once someone understands ( “groks” ) how to approach the Aggregation Pipeline, the power and use of it becomes clear. They can see it does much more than GROUP_BY – they can JOIN data from multiple collections, execute graph traversals, implement faceted search, and more.
This presentation includes a fundamental discussion of the philosophy of the Aggregation Pipeline, how to approach solving problems, and a conceptual approach to designing powerful analytical queries and data transformation statements.