Cray Graph Engine
Uncover valuable insights.
We live in a connected world, and a graph database is about data connections at its core. Viewing related data with a graph database, in addition to relational database tables, uncovers valuable insights that were previously hidden.
That’s why graph analytics is gaining traction in the world of big data. Cray customers are using graph analytics for big data to solve real-world problems such as detecting cyberattacks, creating value from internet of things sensor data and precisely identifying drug interactions faster than ever before. Graph analytics and graph databases have become a powerhouse in detecting never-before-seen connections and emergent patterns.
Cray’s unique approach to graph analytics leverages high-performance hardware and decades of experience with big data and pattern recognition, combined with industry-standard software engineered for sophisticated graph analytics. The result is the Cray Graph Engine (CGE). CGE lets you analyze data using pattern matching and filtering, sophisticated graph algorithms and analysis, in a powerful, interactive system that scales to graphs with billions of edges.
Key benefits of the Cray Graph Engine
- Integrates with classical graph algorithms
- Blazing-fast results
- Industry-standard database structure and query language
Breakthrough insight. Breakout opportunities.
The structure and language of a database shape the type of queries you can run. The Cray Graph Engine is a semantic database using Resource Description Framework (RDF) triples to represent the data, SPARQL as the query language and extensions to support mathematical algorithms.
- RDF — Stores each data item as a triple containing a subject-predicate-object. For example, “John is the parent of Susan”:
- SPARQL — Query language built to access RDF data, syntactically similar to SQL
- Extensions — New operators to equip our graph database system with classical graph algorithms
CGE’s approach of combining graph database structure and mathematical algorithms provide very comprehensive capabilities for ad-hoc, iterative and interactive analytics.
Speed. At scale.
The Cray Graph Engine, running on our Urika-GX agile analytics platform and Cray XC series supercomputers with Urika-XC analytics, is often 1-2 orders of magnitude faster than the competition, running on a similarly-sized system.
Designed to leverage the Cray Aries™ interconnect for better I/O and a high degree of parallelism, the Cray Graph Engine delivers a powerful, interactive graph solution that scales to graphs with tens of billions of edges. And perhaps most important, CGE provides results in minutes — not hours or days.