Advances in the Tau Performance System – Heterogeneity and Databases and Autotuning! Oh My!

Colloq: Speaker: 
Allen D. Malony, PhD
Colloq: Speaker Institution: 
University of Oregon
Colloq: Date and Time: 
Thu, 2012-10-11 14:45
Colloq: Location: 
Building 5700, Room L202
Colloq: Host: 
Jeffrey Vetter
Colloq: Host Email:
Colloq: Abstract: 
The talk will cover recent advancements in the TAU Performance System. Specifically, we will look first at TAU's evolving support of measurement and analysis of heterogeneous platforms with accelerators. Next, the re-engineering of TAU's performance database (TAUdb) will be described and discussed as a foundation for multi-experiment performance investigation. Finally, we will look at recent work in the integration of TAU performance measurement, TAUdb, data mining, and machine learning in auto-tuning frameworks.
Colloq: Speaker Bio: 
Dr. Allen D. Malony is a Professor in the Department of Computer and Information Science at the University of Oregon. He received the Ph.D. degree from the University of Illinois at Urbana-Champaign in 1990. Dr. Malony's research interests are in parallel computing, performance analysis, supercomputing, and scientific software environments. He directs the TAU project at Oregon where his research group develops the TAU Performance System (TM), a leading open source parallel performance tool suite in use by many academic, governmental, and industrial projects around the world. Dr. Malony has been recognized as a NSF National Young Investigator (1994), Fulbright Research Scholar (The Netherlands, 1992; Austria, 1999), and Humboldt Senior Research Scientist (2002). Dr. Malony's research interests extend to computing and information processing systems for neuroscience research. He is the Director of the NeuroInformatics Center (NIC) at the University of Oregon. The NIC is developing advanced integrated neuroimaging tools that combine EEG and MRI methods for next-generation brain analysis. In 2004, Dr. Malony, together with Dr. Sameer Shende, founded ParaTools, Inc., a company providing expertise in parallel software and tools for application development, analysis, and optimization.