Microsoft: Bring clarity to complex data with graph databases on SQL Server 2017
SQL Server 2017, in addition to processing relational data, now fully integrates with graph database models, all on the same familiar system. This will bring clarity to the increasing amounts of data businesses generate every day.
What’s the difference between graph and relational databases?
Relational databases, like SQL Server, use foreign keys to manage relationships between entities and tables. Foreign keys adequately query one-to-many relationships; however, as relationships between various data entities become more complex, queries also become more complex and performance may diminish as a result.
In these cases, developers may opt for graph database models to manage complex relationships and enable operational agility. In a graph database, edges are heterogeneous in nature—a single edge can be used to connect different type of nodes to each other. This is not easy to achieve using foreign keys in a relational database. For example, consider a social graph where a person (node) likes another person (edge) or organization (node) or restaurant (node). Here the same ‘likes’ edge is used to connect three different types of nodes and entities to each other—person to person; person to organization; and person to restaurant.
So, how do graph databases work?
Graph databases are comprised of a collection of nodes and edges. Nodes represent a given entity (a noun), while edges (a verb) represent the relationship between the two nodes they connect. For example, pitcher Randy Johnson (node) faces (edge) batter John Kruk (node).
Read the entire article here, Bring clarity to complex data with graph databases on SQL Server 2017
Via the fine folks at Microsoft.