‘Graph data’ is normally used to describe data that is highly interconnected. Transportation routes, social media friends or connections, organizational charts, genealogical charts, product and customer data, electrical grids and pipelines…the scenarios where you can find them are endless. The most common way graph data is modeled in the relational world is by using many-many-relationships..querying such structures can be very painful and expensive in terms of time and computing power. Modeling it the graph way instead allows for efficient querying, detection of associations and patterns, performing affinity analysis and adding business value in a variety of different ways.
There are dedicated graph databases that do this..such as Neo4j..but what if most of your data is in a relational engine already? What if you want to keep the gains of a relational engine such as security, ACID properties and high availability and still do some graph modeling? SQL Server 2016+ supports graph data modeling as well. In this talk on SQL Graph Revealed we will learn about origins of graph theory, components of graph data, and advantages of modeling relationships using graph capabilities of SQL Server.
Mala Mahadevan is a senior database professional with over 20 years of experience working with data, primarily in SQL Server and related technologies. She has been volunteering with SQL Server community for the past 15 years and is also a recipient of the PASSion award for being an outstanding volunteer. She is a featured blogger on sqlservercentral.com and also blogs frequently at curiousaboutdata.com. She is active on twitter as @sqlmal.