June 7, 2012

Center for Adaptive Supercomputing Software

John Feo

Manager Computational Sciences & Mathematics Division
Adaptive Supercomputer Software Group at Pacific Northwest National

Data collection and analysis is rapidly changing the way scientific, national security, and economic communities operate. American economic competitiveness and security will depend increasingly on the timely analysis of Petabytes data sets. Diverse formats, errors, and uncertainties complicate the analysis process. Unlike traditional scientific applications based on linear algebra routines, data analytic applications comprise large, integer-based graph computations with irregular data access patterns, low computation to memory access ratios, and high levels of fine grain parallelism that pass data and synchronize frequently. Traditional architectures optimized to run large-scale floating point intensive simulations are inadequate, and more suitable highly multithreaded architectures are needed. In this talk I will survey the research at PNNL’s Center for Adaptive Supercomputer Software in graph algorithms, software tools, and next generation multithreaded architecture design. We have active work in network analysis, natural language understanding, cybersecurity, machine learning, community detection, and benchmarks. We are developing a scalable semantic database framework including new query language, search engine, data management, and system support. We are working to improve the ability of open source compiler, runtime, and tools for performance and power to support irregular, multithreaded applications. In the area of system design, we are investigating new hardware configurations, software emulation for multithreading, and evaluating emerging systems.

Presentation view/download PDF

Part 1 Part 2 Part 3 Part 4 Part 5 Part 6

About John Feo

Dr. John Feo is the director of the Center for Adaptive Supercomputer Software at the Pacific Northwest Laboratory. 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. After a short two year “sabbatical” at Microsoft where he led a software group developing a next-generation virtual reality platform, he joined PNNL Dr. Feo’s research interests are parallel programming, graph algorithms, multithreaded architectures, functional languages, and performance studies. He has published extensively in these fields. He has held academic positions at UC Davis and is an adjunct faculty at Washington State University.