Graph databases go back to the early 2000s and have long been seen as a great way to answer questions about complex relationships in large data sets. However, for a long time they struggled to perform well when data volumes grew large and when the answers must be provided in real time.
First-generation graph databases were built with native graph storage but were not made to handle large data or query volumes or perform beyond three levels or connections– known as hops – inside the graph. With every hop in a graph, the scope of the search expands dramatically and the insights gleaned become deeper.
Second-generation graph databases were built on top of NoSQL storage, which allowed them to load large amounts of data. However, they still do not scale for queries involving three or more hops. Older graph databases also typically do not support “database sharding”– partitioning of data across a number of servers to increase scalability-which means, a large graph with terabytes of data can’t be distributed. These legacy graph databases are ill-equipped to scale up to today’s real-world requirements, which call for a system that can perform many hops efficiently and in parallel to deliver sub-second query performance on big data.
TigerGraph is a new kind of graph database, a native parallel graph database purpose-built for loading massive amounts of data (terabytes) in hours and analyzing as many as 10 or more hops deep in to relationships in real-time. TigerGraph supports transaction alas well as analytical workloads, is ACID compliant, scales up and out with database sharding. TigerGraph’s proven technology supports applications such as fraud detection, customer360, IoT, AI and machine learning to make sense of ever-changing big data, and is used by customers including Visa, Intuit, China Mobile, Wish and Zillow.