SYSTAP, LLC, developer of the leading edge bigdata® RDF graph database platform, today announced receiving research funding totaling $2M from the Defense Advanced Research Projects Agency (DARPA) to develop an open source platform for accelerated and real-time graph analytics on GPU compute clusters. This contract is part of DARPA’s XDATA program, a 4-year research effort to develop new computational techniques and open source software tools for processing and analyzing data, motivated by defense needs. SYSTAP, LLC has been selected by DARPA as a performer in the technical area of scalable analytics and data processing technology. The contract is administered by the Air Force Research Laboratory, Rome, NY.
Graphs are everywhere in today’s society, in social networks, business transactions, the structure of the Internet, the structure of ecosystems and markets, the nature of the genome and cell processes. Graphs take us beyond tables and relational models, making it easy for people to mash up data from different sources. Graph processing algorithms help people turn complex, uncertain, and messy data into actionable information. By combining graphs, GPUs (Graphical Processing Units), and techniques from High Performance Computing (HPC), we hope to achieve a new capability for scientific discovery and real time business analytics.
GPUs and graphs are the tipping point. Together, they will let us address the largest problems in science and business, but the commodity price point of GPUs means that even the smallest business can have the compute power of a 30-node cluster in a $10k workstation. The challenge is to give people the software tools that make it easy to apply this compute power to their data. By combining open data and linked data with high performance graph processing, people will be able to create value new services from existing data.
SYSTAP, LLC develops innovative, scalable open source platforms for graph databases and graph data mining. SYSTAP, LLC leads the development of the bigdata® RDF graph database platform and offers support subscriptions, training, and custom services for the platform. The bigdata® platform is also available under commercial licenses from a number of OEMs and VARs.
Contact: Bryan Thompson
University of California, Davis – Dr. John Owens. Dr. Owens will lead the development of an open-source, scalable, multi-node, out-of-core graph library on GPUs. Dr. Owens is an Associate Professor of Electrical and Computer Engineering. He leads a research group pursuing problems in GPU computing in both GPU fundamentals (data structures, algorithms, and multi-GPU computing) and applications (including computer vision, GPU-based embedded systems, real-time and offline graphics, medical imaging, speech recognition, protein folding, computational fluid dynamics, and visualization).
Contact: John Owens
University of California, Santa Barbara – Dr. Xifeng Yan. Dr. Yan will provide consultation on basic primitives behind graph data mining, graph query/search, and graph OLAP. He is an Associate Professor and holds the Venkatesh Narayanamurti Chair in Computer Science where he works on modeling, managing, and mining graphs in bioinformatics, social networks, information networks, and computer systems.
Contact: Xifeng Yan
smartRealm specializes in geo-social analytics, linked data, knowledge representation, semantic reasoning and data-knowledge integration. Our flagship product, knowledgeSmarts® (kS), is an award-winning Knowledge-as-a-Service™ (KaaS) platform, and features kS Workbench to knowledge-enable enterprise environments.
Contact: Stephane Fellah, CTO and Harry Niedzwiadek, General Manager. 703-669-5514.
For more information on DARPA and the XDATA program, visit www.darpa.mil
The views expressed are those of the author and do not reflect the official policy or position of the Department of Defense or the U.S. Government.
Approved for Public Release, Distribution Unlimited.