Tables, Graphs, and Problems
DEIB - Seminar Room
May 19th, 2016
2.30 pm
John Feo
Director of the Northwest Institute for Advanced Computing, Washington
Contact:
Gianluca Palermo
Research Line:
System Architectures
May 19th, 2016
2.30 pm
John Feo
Director of the Northwest Institute for Advanced Computing, Washington
Contact:
Gianluca Palermo
Research Line:
System Architectures
Sommario
The availability of data is changing the way science, business, and law enforcement operate. Economic competitiveness and national security depend increasingly on the insightful analysis of large data sets. The volume, variety, velocity, and varacity of data sets and the breadth of analytic processes pose unique computational challenges. Traditional table-based methods are being supplemented by graph methods that operate on sparse data capturing dynamic relationships among typed entities. Consequently, analytic platforms must support both in a natural way without preference. In this talk, I will summarize some of the emerging trends in data analytics focusing on the hybrid nature of big data, the consquences of scale-free real world data sets, and essential runtime requirements for scable performance. I will describe how we are addressing these challenges in GEMS 2.0, a second-generation database platform. The first generation database was pure graph --- RDF triples, SPARQL queries, and constrained graph isomorphism. We found it inadequate for many real-world problems that comprise a significant amount of dense data and table operations. So, we redesigned the database to support both tables and property graphs, and have extended SQL to define graph views and queries. GEMS 2.0 sits atop a custom, fine-grain, multithreaded runtime system that scales on conventional server and HPC systems.
Biografia
Dr. John Feo is the Director of the Northwest Institute for Advanced Computing, a joint institute established by Pacific Northwest National Laboratory and University of Washington. Previously, he managed a large DOD research project in graph algorithms, search, parallel computing, and multithreaded architectures. Dr. Feo received his Ph.D. in Computer Science from The University of Texas at Austin. He began his career at Lawrence Livermore National Laboratory where he managed the Computer Science Group and was the principal investigator of the Sisal Language Project. Dr. Feo then joined Tera Computer Company (now Cray Inc) where he was a principal engineer and product manager for the MTA-1 and MTA-2, the first two generations of the Cray’s multithreaded architecture. He has taken short sabbaticals to work at Sun Microsystem, Microsoft, and Context Relevant. Dr. Feo’s has held academic positions at UC Davis and Washington State University.