Titan
Titan is a distributed and scalable graph database built on Cassandra, Hbase, and Elasticsearch. Based on my PhD research at the University of Maryland, Titan pioneered many of the methods and techniques for scaling graph data under transactional workloads. I led the development of Titan as CTO of Aurelius. No longer under development - replaced by JanusGraph.
Visit Project
DataStax Enterprise Graph
DSE Graph is a distributed graph database optimized for enterprise applications - zero downtime, fast traversals at scale, and real-time analysis of complex, related datasets. DSE Graph is the evolution of Titan as an enterprise-grade database system. I led the development of DSE Graph after the acquisition of Aurelius by DataStax.
Visit Project
Apache TinkerPop
Apache TinkerPop is a graph computing framework for both graph databases and graph analytic systems. The popular graph query language Gremlin is part of TinkerPop. Apache TinkerPop is an integral part of both Titan and DSE Graph.
Visit Project
AppStax
A research project and product prototype aimed at greatly simplifying database application development. Users can define their data model and queries and AppStax generates the optimal, database-backed REST, GraphQL, or gRPC API for them. No longer under development - replaced by Stargate.io.
Visit Project
Probabilistic Soft Logic
Probabilistic Soft Logic is a machine learning framework for developing probabilistic graphical models. Compared to other SRL frameworks, PSL is easy to use and provides fast inference. I invented PSL as part of my PhD at the University of Maryland. PSL is developed by the LINQS research group at the University of California, Santa Cruz.
Visit Project
COSI
COSI introduces a number of novel techniques for distributing graph data across a cluster of compute machines in such a way that queries can be answered efficiently. Some of these ideas were later implemented in the Titan graph database.
Visit Project
Social Network Data Management
My PhD thesis in Computer Science at the University of Maryland. The thesis ties together multiple lines of research centered around graph database management systems and graph structured machine learning.
Visit Project
DOGMA
DOGMA proposes novel indexing structures to more efficiently represent graph structured data on disk by clustering related data on disk pages for faster seek times.
Visit Project