Graph processing in query languages using UDAFs

Introduction Processing graphs is a fundamental problem area that has applications in several domains ex: networks. One of the building-block problems in graph processing is “finding connected components (sets of fully connected nodes)”. It is not straight forward to express graph problems as structured queries unless the database itself exposes primitives for graph processing ex: graph oriented databases like neo4j might do this . In this article I will attempt to present that graph processing problems can be expressed as queries as long as the database (or the datawarehouse) allows extending its core functionality through UDAF(User Defined Aggregation Function). I […]