Real-Time Deep Link Analytics for BigGraph: Challenges and Solutions

Join us for tapas, sangria and an educational talk on Big Graph on April 11:

"Real-Time Deep Link Analytics for BigGraph: Challenges and Solutions"

More and more organizations are seeing the value that graph databases can provide: a data model that naturally expresses the relationships between entities, and the ability to define better predictive models and gain deeper insight by analyzing the graph. However, the ability to scale to the massive size needed for some applications while delivering the required performance has been a challenge for graph databases such as Neo4j in the past.

Key performance metrics include the following:

• Query latency and query throughput, for queries requiring 2, 3, or more hops.
• Data loading and data update speed
• Performance vs. scale out, partitioning the data across multiple machine nodes.

In this talk, we review several use cases for graph databases and their performance requirements. We also consider the factors that have tended to inhibit performance on graph databases. We present TigerGraph whose native parallel graph design offer a scalable, real-time, deep-link graph analytics. We provide some benchmark results comparing TigerGraph to Neo4j and Titan.

5:30 PM - Networking and drinks
6:00 PM - Food and Presentation
7:00 PM - More networking

SPEAKER:
Dr. Victor Lee is Director of Product Management at TigerGraph. His first stint in Silicon Valley was as an IC circuit designer and technology transfer manager, before returning to school for his computer science PhD, focusing on graph data mining. He received his BS in Electrical Engineering and Computer Science from UC Berkeley, MS in Electrical Engineering from Stanford University, and PhD in Computer Science from Kent State University.