InTheLoop | 05.21.2012
May 21, 2012
Turning Sunlight into Electricity Like a Plant
Solar power remains more expensive on average than fossil fuels; one reason is that traditional photovoltaics require expensive rare-earth elements. If we learn from plants, which use only common elements—hydrogen, nitrogen, carbon, oxygen and some others—to convert sunlight into energy, then we’ll be able to bring down the cost of solar power. This is why researchers are looking at bio-inspired materials as possible resources for solar energy.
In the 1990s, an Arizona State University research group made a huge advance in this field by creating the carotenoid-porphyrin-C60 molecular triad, a novel material that converts sunlight into chemical energy by mimicking photosynthesis. However, the material has been difficult to commercialize because it can only be controlled, or confined, in experimental labs.
But now, using 2 million computer hours at NERSC and 2.5 million computer hours at the Texas Advanced Computing Center (TACC), University of Houston physicist Margaret Cheung and her team have explored the role that confinement, temperature, and solvents play in the stability and energy efficiency of the light-harvesting triad. Their results provide a way to test, tailor, and engineer nano-capsules with embedded triads that, when combined in large numbers, could greatly increase the ability to produce clean energy. Read more.
Turning Water into Hydrogen Fuel Using a More Reactive Catalytic Surface
Although fuel cells have been touted as a clean alternative to combustion engines for powering cars, the molecular hydrogen required to fuel this technology is naturally rare on Earth and must be extracted from natural gas or water. To split water molecules, the popular catalyst titanium dioxide (TiO2) needs an even layer of hydroxyl (OH) groups across its surface. Conventional methods for putting hydroxyl groups on TiO2 achieve about 20 percent coverage, but scientists have been trying to beat those results, getting more coverage without resorting to extremes of time, temperatures, or resources.
Now a team of researchers at the Pacific Northwest National Laboratory (PNNL) and the Worcester Polytechnic Institute has figured out how to cover 50 percent of a TiO2 surface with hydroxyl groups. Using resources at NERSC and the Environmental Molecular Science Laboratory (EMSL) at PNNL, the team also characterized the atomic-level structure and reactivity of the hydroxyl-rich TiO2 surface. Read more.
Director’s Awards for Exceptional Achievement Announced
In October 2011, Berkeley Lab launched a new annual awards program to honor exceptional achievements by laboratory employees advancing the Lab’s mission and strategic goals, and the first winners were announced today. Honorees will be acknowledged at a future Lab awards ceremony and reception. One of the Operations awards went to the Green Datacenter Team, which includes ESnet’s Acting Division Director Greg Bell. See the complete list of winners.
New LBNL Software Developers Group Already Has 100 Members
Deb Agarwal of CRD, Eric Pouyoul of ESnet, David Skinner of NERSC, and Adam Stone of the IT Division are forming a new group for software developers at Berkeley Lab named firstname.lastname@example.org. The purpose of the group is to connect up the software developers throughout the Lab via an email list, a Google group, and seminars/brown bags on topics of interest to coders. It’s a place for programmers to ask advice from fellow programmers, share successes, and discuss software engineering issues. If you are interested in joining the list, visit https://groups.google.com/a/lbl.gov/group/coders/ to sign up. They already have over 100 members and are working on arranging the first events.
CS Staff Participating in IPDPS 2012 in Shanghai
The 26th IEEE International Parallel and Distributed Processing Symposium (IPDPS 2012) is being held May 21–25 in Shanghai, China. Associate Lab Director Kathy Yelick will moderate a panel discussion on “Will exascale computing really require new algorithms and programming models?” Papers with co-authors from CRD include “New Scheduling Strategies for a Parallel Right-Looking Sparse LU Factorization Algorithm on Multicore Clusters” (Sherry Li and former staffer Ichitaro Yamazaki) and “PGAS for Distributed Numerical Python Targeting Multi-Core Clusters” (Yili Zheng).
Horst Simon Testifies before Congress on China Computing Efforts
On May 10, Deputy Lab Director Horst Simon was in Washington, D.C. providing testimony on China’s high performance computing program before a hearing of the U.S.-China Economic and Security Review Commission. Go here to view a video of the hearing, entitled “Assessing China’s Efforts to Become an Innovation Society.” Congress created the 12-member commission in 2000 to study and to provide recommendations on the bilateral security and economic relations between the United States and China.
CRD’s Phil Colella Named Chair of Lab Staff Committee
Phil Colella, a Senior Scientist in the Computational Research Division, has been named as chair of the Laboratory Staff Committee (LSC). He replaces Physics Division Director Natalie Roe. The LSC advises the Lab director on matters concerning Lab-wide consistency in the quality of the scientific staff and on divisional procedures for selection and appointment. Read more.
Yeen Mankin to Take New Position Supporting ALS Director Roger Falcone
Yeen Mankin, who has provided support for Computing Sciences meetings and conferences, including Berkeley Lab’s annual exhibit at the SC conference series, has accepted a position as senior administrator for Roger Falcone, director of the Advanced Light Source Division. Her last day with Computing Sciences will be Thursday, May 31.
“Yeen’s departure is a real loss for us—and a great gain for the ALS,” said Associate Lab Director Kathy Yelick. “We have come to rely on her expertise, good judgment and seamless execution of her job, no matter what challenges came her way. We’ll miss her and wish her all the best.”
Mankin joined the Lab in late 2000 from CBS MarketWatch, where she was an executive administrator. She had also worked for public relations consultancies in her native Malaysia.
Funding Available to Attend Grace Hopper Conference
The Grace Hopper Celebration of Women in Computing conference will be held in Baltimore, MD October 3-6, 2012. This conference is designed to bring the research and career interests of women in computing to the forefront. Presenters are leaders in their respective fields, representing industrial, academic, and government communities. Leading researchers present their current work, while special sessions focus on the role of women in today’s technology fields, including computer science, information technology, research, and engineering.
The Berkeley Lab Computing Sciences Diversity Working Group is able to pay the travel and registration of a few Lab staff to attend. If you would like to attend the Grace Hopper Conference and would like to have your travel costs sponsored by the Diversity Working Group, please contact Elizabeth Bautista. Send the following information:
- Paragraph giving the reason you would like to attend Grace Hopper
The deadline for submitting applications is June 1, 2012.
This Week’s Computing Sciences Seminars
HP and High-Performance Computing at Cal
Tuesday, May 22, 10:00 am–1:00 pm, Wozniak Lounge, Soda Hall, UC Berkeley
HP and its high-performance computing partners like ScaleMP, Mellanox, Nvidia, and DDN will give presentations on industry-leading solutions. Join us for this informative and casual event to learn how HP and its partners can help you achieve your research objectives, leveraging today’s technology solutions with an eye toward what is coming next.
Carbon Cycle 2.0 Seminar: Mining Metagenomic Data for New Enzymes
Thursday, May 24, 2:00–3:00 pm, 15-253
Zhong Wang, LBNL Genomics Division/JGI
The success of cellulosic biofuels and artificial photosynthesis both rely on the discovery of diverse and efficient enzymes. Recent breakthroughs in deep metagenome sequencing can produce millions of genes from a single experiment, thereby providing a large pool for identifying new enzymes. Here I will present our progress in developing informatics solutions for enzyme prediction from metagenomic data. In the first part of my talk I will discuss a homology based strategy that identified a large number of cellulases from the cow rumen biomass degradation community; in the second part, I will discuss a machine learning based approach that is homology independent. Called Buchner, this approach is very computationally efficient, and capable of discovering distant homologs or structural homologs that more likely represent novel enzymes.
Relativistic Explicit Correlation
Friday, May 25, 11:00 am–12:00 pm, 50B-2222
Sihong Shao, Peking University, China
The electron-electron coalescence conditions are derived for the wave functions of the Dirac-Coulomb (DC), Dirac-Coulomb-Gaunt (DCG), Dirac-Coulomb-Breit (DCB), modified Dirac-Coulomb (MDC), and zeroth-order regularly approximated (ZORA) Hamiltonians. The manipulations make full use of the internal symmetries of the reduced two-electron Hamiltonians such that the asymptotic behaviors of the wave functions emerge naturally. The results show that, at the coalescence point of two electrons, the wave functions of the DCG Hamiltonian are regular, while those of the DC and DCB Hamiltonians have weak singularities. The behaviors of the MDC wave functions are related to the original ones in a simple manner, while the spin-free counterparts are somewhat different due to the complicated electron-electron interaction. The behaviors of the ZORA wave functions depend on the chosen potential within the kinetic energy operator. In the case of the nuclear attraction, the behaviors of the ZORA wave functions are very similar to those of the non-relativistic ones, just with an additional correction to the non-relativistic cusp condition. However, if the Coulomb interaction is also included, the ZORA wave functions become close to the large-large components of the DC wave functions. Note that such asymptotic expansions of the relativistic wave functions are only valid within an extremely small convergence radius. Beyond this radius, the behaviors of the relativistic wave functions are still dominated by the non-relativistic limit, as can be seen in terms of direct perturbation theory (DPT) of relativity. Another deduction is that the DC Hamiltonian has no bound electronic states, although the last word cannot be said. These findings enrich our understandings of relativistic wave functions. This is a joint work with Zhendong Li and Wenjian Liu.
Link of the Week: Zettaflops Will Happen, Says HPC Analyst
“While Thomas Sterling’s interview about the impossibility of reaching zettaflops made a lot of sense, the history of making negative predictions about technology is often an embarrassing one,” writes John Barr, Research Director High Performance Computing, 451 Research, in HPCwire.
“Professor Sterling pioneered the use of compute clusters and is a Gordon Bell Prize winner,” Barr continues. “He has excellent credentials in HPC, and I can’t refute a single fact that he put forward in his interview—indeed, I am generally in full agreement with [his] insights on the issues the industry faces—but I am certain that he is wrong in his conclusion.” Read more.
About Computing Sciences at Berkeley Lab
The Lawrence Berkeley National Laboratory (Berkeley Lab) Computing Sciences organization provides the computing and networking resources and expertise critical to advancing the Department of Energy's research missions: developing new energy sources, improving energy efficiency, developing new materials and increasing our understanding of ourselves, our world and our universe.
ESnet, the Energy Sciences Network, provides the high-bandwidth, reliable connections that link scientists at 40 DOE research sites to each other and to experimental facilities and supercomputing centers around the country. The National Energy Research Scientific Computing Center (NERSC) powers the discoveries of 6,000 scientists at national laboratories and universities, including those at Berkeley Lab's Computational Research Division (CRD). CRD conducts research and development in mathematical modeling and simulation, algorithm design, data storage, management and analysis, computer system architecture and high-performance software implementation. NERSC and ESnet are DOE Office of Science User Facilities.
Lawrence Berkeley National Laboratory addresses the world's most urgent scientific challenges by advancing sustainable energy, protecting human health, creating new materials, and revealing the origin and fate of the universe. Founded in 1931, Berkeley Lab's scientific expertise has been recognized with 13 Nobel prizes. The University of California manages Berkeley Lab for the DOE’s Office of Science.
DOE’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.