TECHEX EXCHANGE SPEAKER TO SHARE EMERGING "SUPERFACILITY" SCIENCE DISCOVERY MODEL
Prof. Kathy Yelick, Associate Laboratory Director (ALD) for Computing Sciences at Lawrence Berkeley National Laboratory, will give the TechEX17 Exchange address, "The Superfacility: A New Supermodel for Seamless Scientific Discovery," on Monday, October 16.
Yelick's talk will describe the Superfacility model - a new paradigm for scientific research that combines HPC, networking and experimental facilities into a holistic solution. The talk will cover early success stories in areas of biology, astrophysics and others; and the current gaps in technology (workflow tools, algorithms for analysis, programming models, data storage, and data security, etc.) that need to be addressed to move the Superfacility model from a pilot effort to a production and scalable service.
As the Berkeley Lab Computing Sciences ALD, Yelick oversees the Energy Sciences Network (ESnet), the National Energy Research Scientific Computing Center (NERSC), and the Computational Research Division which performs research in applied math and computer science and develops tools and techniques for advancing scientific discovery. Together, they form an ecosystem for science, which laid the groundwork for the Superfacility concept.
Yelick is also a professor of electrical engineering and computer sciences at the University of California, Berkeley. Her research is in parallel programming languages, compilers, algorithms, and automatic performance tuning. Yelick was director of NERSC from 2008-12. Yelick was recently elected to the National Academy of Engineering (NAE) and the American Associate of Arts and Sciences, and is an ACM Fellow and recipient of the ACM/IEEE Ken Kennedy and Athena awards.
Presentation Abstract: The ubiquity of smartphones, sophisticated cloud services and the rapid pace of innovative technology entering the consumer market every day has created extreme misperceptions about how science discovery works. Even as technology becomes more pervasive and adds layers of complexity to everything we do, scientists and engineers are working furiously behind the scenes to build simple, scalable solutions. When it comes to big science, weaving together the most powerful experimental facilities, the fastest R&E networks and world-leading supercomputers, smoothing the way is one of the largest challenges facing the research community. Rising to the challenge, Lawrence Berkeley National Lab is piloting a revolutionary model which delivers the power of HPC to the experiment site via high-performance networking such that these operate as a single seamless instrument. This new Superfacility model is poised to transform experimental science and deliver on the promise of the data age by making possible discoveries that would otherwise be out of reach.