InTheLoop | 04.20.2015
Special Journal Issue Spotlights NERSC
The IEEE/AIP magazine Computing in Science and Engineering has published a special issue devoted to NERSC, which celebrated its 40th anniversary in 2014. In their introduction, guest editors Harvey Wasserman and Richard Gerber of NERSC outline the history of the center and write, "as the first unclassified supercomputer, NERSC was, and has become, the model for those that followed." Julian Borrill, Reijo Keskitalo and Theodore Kisner of the Computational Cosmology Center at Berkeley Lab review 15 years of exploring the origins and fate of the universe via the cosmic background radiation at NERSC. NERSC's Jack Deslippe, Brian Austin, Chris Daley and Woo-Sun Yang write about the challenges the center faces in programming for future systems. NERSC's Yushu Yao and co-authors delve into dealing with the enormous, structured data sets that increasingly drive scientific discovery. And fusion energy researchers from the Princeton Plasma Physics Laboratory explore NERSC's long (and distinguished) role in this area. At its 1974 founding, the center was dedicated solely to fusion energy research. »Read more.
James Demmel Named ACM Computing Innovator
James Demmel, a professor at the University of California, Berkeley, who has a joint appointment in Lawrence Berkeley National Laboratory’s Computational Research Division, received the Association for Computing Machinery’s (ACM’s) Paris Kanellakis Theory and Practice Award. Demmel was honored for his work on numerical linear algebra libraries, including LAPACK (Linear Algebra Package), a standard software library that forms part of the standard mathematical libraries for many vendors.
According to ACM, the Paris Kanellakis Theory and Practice Award honors specific theoretical accomplishments that have had a significant and demonstrable effect on the practice of computing. »Read more.
Electrolyte Genome Could Be Battery Game-Changer
The Materials Project hosted at NERSC is taking some of the guesswork out of discovering new battery catalysts with a project called the Electrolyte Genome. “Electrolytes are a stumbling block for many battery technologies, whether the platform is designed for electric vehicles or a flow battery for grid applications,” said Kristin Persson, a Berkeley Lab scientist who serves as the NERSC principal investigator on the Materials Project. “What we can do is calculate the properties of a large number of molecules and give experimentalists a much better set of materials to work with than if they were to explore all possible combinations.”
She explained: “If we can come up with an electrolyte that has a higher electrochemical window for multivalent batteries, or with larger solubility for certain redox molecules, if we can solve either of these, you suddenly enable the whole industry. It could be a game-changer.” »Read more.
National Geographic Quotes Wehner on Extreme Weather
In a recent story about extreme weather and climate change, the National Geographic turned to the lab's Michael Wehner, a computational climatologist with the Computational Research Division. Extreme weather changes, said Wehner, will likely become noticeable with a rise of 2°C in the average global surface temperature.“Definitely, at two degrees, there will be large changes to the number of Category 4 and Category 5 storms, and the very strongest storms will be stronger,” he says. “You often read that two degrees is some arbitrarily safe form of climate change. I take issue with that." »Read more.
This Week's CS Seminars
Nonequilibrium processes at the molecular level: an update on two conventional "diffusion pictures" by computational modeling
Monday, April 20, 10:00am - 11:00am, Bldg. 50B, Room 4205
Nils Zimmerman, Computational Chemistry, Materials & Climate Group, Lawrence Berkeley National Laboratory
Diffusion is decisive in applications such as gas separation, heterogeneous catalysis, and crystal growth. In my talk, I showcase two situations where insights from nonequilibrium molecular simulations have changed the view of how diffusion affects the systems studied. The first part of the talk addresses transport of gas molecules into the nanopores of zeolites. For these systems, Fick’s laws have been shown to be reliable if large crystals are used–both in experiments and simulations. However, if molecules enter thin zeolite nanosheets the underlying assumptions of Fick’s laws become questionable. This issue is investigated by using transient molecular dynamics simulations, a computational modeling approach that closely mimics situations encountered in diffusion experiments. Results for methane entering two different zeolite types highlight that Fick’s laws break down for small host sizes and smooth pores. This is because kinetic jump correlations violate random walk theory in such cases. In the second part, I will present a case study of sodium chloride (rocksalt) nucleation from aqueous solution. The seemingly simple system is chosen because nucleation still poses notorious challenges to theory, simulations, and experiments. For example, it has recently been shown by metadynamics simulations that rocksalt might form via a transient wurtzite-like polymorph. We developed local order parameters that can distinguish between the two structures and found that this was a model artifact. Using a reliable model, we calculated the nucleation rate of NaCl in water on the basis of classical nucleation theory and a seeded simulation approach. The analysis of attachment to the critical nucleus revealed that the kinetics are not limited by diffusion. This is however the typical assumption for predicting the attachment frequency. Finally, our rates hint at the difficulty of identifying the true time when the very small critical nucleus forms in experiments.
CITRIS Seminar–Managing Water Scarcity for Future Cities:
What Makes Sense to Sense?
Wednesday, April 22, 12pm - 1pm, Banatao Auditorium, 310 Sutardja Dai Hall, UC Berkeley
Erik Porse, California Center for Sustainable Communities and the Institute of the Environment and Sustainability
Urban water infrastructure systems must meet future challenges in managing water quantity and quality in growing cities worldwide. Information and monitoring technologies will play an important role. Within engineering and planning fields, resilient systems typically support quick returns to normal operations following outages. Other conceptions of resilience from ecology, however, recognize that current systems can undergo rapid and potentially unpredictable changes. This conception becomes increasingly important for cities facing evolving risks of extreme rainfall, flooding, or water scarcity. Metropolitan Los Angeles is promoting a significant transition to greater local reliance for water supplies, using stormwater capture, recycled water, and groundwater banking to reduce reliance on imported water sources. New information technology and monitoring capabilities will help fill information gaps created by the region's institutionalized system of water governance. For instance, new information on localized infiltration and water quality for stormwater and groundwater basins can support changes in regulations, the urban landscape, and water recycling. I will discuss on-going research within the California Center for Sustainable Communities demonstrating the importance of detailed data for creating resilient and sustainable future cities. »Free registration and more information.