InTheLoop | 06.20.2011
June 20, 2011
NERSC’s Hopper Is Among World’s Top 10 Fastest Computers
The 37th edition of the Top500 List was announced this morning at the 2011 International Supercomputing Conference in Hamburg, Germany. For the first time, all of the top 10 systems achieved petaflop/s performance — and those are also the only petaflop/s systems on the list. With a Linpack benchmark performance of 1.054 petaflops, Hopper ranks eighth in the world and fourth in the United States. Read more.
NERSC Honored with HPC Innovations Excellence Award
This morning at the ISC '11 in Hamburg, Germany, NERSC was honored with an HPC Innovation Excellence Award for providing supercomputing, storage, and service support to the 20th Century Reanalysis
Project—a collaboration of the University of Colorado, National Oceanic and Atmospheric Administration, the Atmospheric Circulation Reconstructions over the Earth Initiative, and 30 international organizations.
Led by University of Colorado climatologists Gilbert Compo and Prashant Sardeshmukh, and NOAA meteorologist Jeffrey Whitaker, the Project uses supercomputers to reconstruct global historical weather maps from 1871 to present day. This dataset helps the science community put current weather extremes in a historical perspective, determine how extremes are changing, and validate computer climate models.The award was created by International Data Corporation (IDC) to recognize noteworthy achievements by users of high performance computing technologies. Read more.
David Leinweber Named One of Top 10 Innovators of Decade in Trading Industry
David Leinweber of the Computational Research Division has been named by Advanced Trading magazine as one of its “Top 10 Innovators of the Decade” for his work in developing a service that allows trading strategies to react to news the instant it breaks, managing what the magazine describes as “a fire hose of aggregated updates.”
Last year, Leinweber joined Berkeley Lab from UC Berkeley and established the Center for Innovative Financial Technology (CIFT) to help build a bridge between the computational science and financial markets communities. Leinweber, who has a Ph.D. in applied mathematics from Harvard University, may be best known as the author of Nerds on Wall Street: Math Machines and Wired Markets, published in 2009. Read more.
Computing Sciences to Host I3P Meeting This Week
The Institute for Information Infrastructure Protection (I3P), a consortium of leading universities, national laboratories and nonprofit institutions dedicated to strengthening the cyber infrastructure of the United States, will hold its next meeting June 22-23 at Berkeley Lab. Deb Agarwal and Sean Peisert are the institutional representatives for the I3P, which rotates its meetings between consortium members. Among I3P's activities is a research fellowship program. Peisert was an I3P Fellow in 2007 and Robin Sommer in Agarwal's group was a 2006 fellow. Sean Whalen, who is currently a post-doc working for Peisert, is a 2010-2011 fellow.
ACS Researchers Start Uncertainty Quantification Study Group
Mathematical models intended for computational simulation of complex real-world processes are a crucial ingredient in virtually every aspect of DOE science. Utilization of such computer models requires addressing uncertainty quantification (UQ), a greatly expanding field focusing on systematic uncertainties such as model limitations, sparse inputs, and initial conditions; and statistical uncertainties such as noisy or incomplete data. Within LBNL and CRD in particular, we have an excellent mix of skills in mathematics, modeling, parallel programming, and systems to address these challenges head-on. But we currently have no coherent approach to UQ, in part because the problem space is large and ill-defined: definitions such as “the end-to-end study of the reliability of scientific inferences” (2009 DOE Grand Challenge Report) and “developing rigorous methods to characterize the impact of limited knowledge on quantities of interest” (Stanford UQ group) leave much to the imagination. Which aspects are most important for DOE exascale computing, and why?
To answer these questions, Dan Gunter and Lavanya Ramakrishnan of the Advanced Computing for Science (ACS) Department in CRD are forming an Uncertainty Quantification Study Group that will help familiarize participants with the varied aspects of the UQ challenge, and spur discussions of how CRD research could help address these. This will be an opportunity to cross-germinate ideas and explore collaborations across groups. The proposed format is a weekly meeting, led by one participant and focused around a paper (which all participants should read) on a particular topic. Note: Students are welcome, and not every participant is expected to lead a discussion, but this is not a lecture series; no slides, just thought and discussion.
The first organizational meeting will be held today (June 20) at 10:00 am in 50F-1647. The web site is https://sites.google.com/a/lbl.gov/uq-summer-2011/.
This Week’s Computing Sciences Seminars
Uncertainty Quantification Study Group
Monday, June 20, 10:00–11:00 am, 50F-1647
Dan Gunter and Lavanya Ramakrishnan, CRD
OSF HPC Seminar Series: ESnet — Science Perspective and Network Services
Tuesday, June 21, 12:00–1:00 pm, 943-238
Eli Dart, ESnet
Eli Dart of the ESnet Network Engineering Group will discuss the services that ESnet provides for sites and scientists. The discussion will also include examples of the ways in which scientists use networks today, and how this is likely to change in the future.
What Is NERSC and How Can Their Computer Resources Help You?
Wednesday, June 22, 12:00–1:00 pm, 90-3075
Noel Keen, CRD; Richard Gerber, NERSC; Wangda Zuo, EETD
NERSC has a large amount of computer resources if your code can take advantage of them. How can their computer resources help you? If you find yourself waiting several hours for your computer to perform a calculation or could envision an interesting result if your computer was 1000x faster, you might find resources at NERSC helpful.
- There are many software tools available, including various compilers.
- Many instances of serial calculations can be executed at the same time in a cluster environment.
- Take advantage of modern multicore processors or GPUs using threading.
- Take advantage of the large number of cores and memory using message passing (MPI) or some other parallel paradigm.
One example of using NERSC for research in EETD is Noel’s work:
- Improve the code performance of EnergyPlus using OpenMP threads. Testing many different inputs and configurations on NERSC machines.
- Use a parallel version of the Particle Swarm Optimization method to find the best set of EnergyPlus input parameters to fit measured results.
Another example is Wangda’s research on accelerating the annual daylighting simulation of Radiance. He is currently optimizing the program and running it in parallel on GPU using OpenCL. He is interested in using NERSC’s GPU cluster to further accelerate the simulation for parametric studies.
Link of the Week: Obesity—Not Aging—Balloons Health Care Costs
Contrary to popular belief, people who live an unusually long time tend to be healthier during their later years than shorter-lived people. That means longer-lived ones typically have lower medical costs during their golden years. This health dividend more than offsets the health care costs they accrue by outliving less healthy people.
Unfortunately, there’s a giant exception to the rule that the longer life tends to be a healthier one: Obese people are living longer, thanks to factors such as cholesterol-cutting medicines, but much of their extra time is spent in ill health, and as a result, their annual medical bills are some 42 percent higher than those of normal-weight people. In fact, the obesity epidemic has greatly increased the prevalence of chronic diseases such as diabetes; but contrary to much of the media coverage on the epidemic, it has had little effect on mortality rates. As the title of one study put it, “Smoking kills, obesity disables.” 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.