InTheLoop | 10.08.2012
October 8, 2012
Sudip Dosanjh Takes Office as NERSC Director
Sudip Dosanjh officially took office as NERSC Director on October 4, earlier than originally planned. His appointment was announced in August.
State Department’s TechWomen Visit Berkeley Lab
Berkeley Lab researchers in the Computational Research, Physics, and Physical Biosciences divisions hosted TechWomen—from Algeria, Lebanon, and Tunisia—during the month of September. Launched by Secretary of State Hillary Rodham Clinton in 2011, TechWomen is an international exchange that uses technology as a means to empower women and girls in the Middle East and North Africa. Read more.
John Bell Gives UC Berkeley Math Department Colloquium
John Bell, head of the Mathematics and Computational Sciences Department in CRD, spoke at UC Berkeley’s Math Department colloquium on Thursday, October 4. His talk was titled “Between Kinetic Theory and Navier Stokes — Modeling Fluids at the Mesoscale.” In this talk he discussed recent work on the development and analysis of finite-volume methods for solving the equations of fluctuating hydrodynamics for miscible fluid mixtures.
CRD Staff Contribute to VisWeek and LDAV
IEEE VisWeek 2012, the premier forum for advances in scientific and information visualization, is being held next week, .October 14–19, in Seattle, WA. Gunther Weber of CRD is on the program committee. CRD staff contributions include:
- Daniela Ushizima, Dmitriy Morozov, Gunther Weber, Andrea Bianchi, James Sethian: Augmented Topological Descriptors of Pore Networks for Material Science
- Joerg Meyer, E. Wes Bethel, Jennifer Horsman, Susan Hubbard (Earth Sciences Division), Harinarayan Krishnan, Alexandru Romosan: Visual Data Analysis as an Integral Part of Environmental Management
- Oliver Rubel, Cameron Geddes (Accelerator and Fusion Research Division), Min Chen, Estelle Cormier-Michel, E. Wes Bethel: Query-Driven Analysis of Plasma-Based Particle Acceleration Data (poster)
- Gunther Weber and collaborators: Interactive Visual Analysis of Scientific Data (tutorial)
The IEEE Symposium on Large-Scale Data Analysis and Visualization (LDAV) 2012
is being held concurrently (October 14–15) in Seattle. Wes Bethel and Hank Childs are on the program committee. CRD staff contributions include:
- David Camp: Parallel Stream Surface Computation for Large Data Sets
- Gunther Weber: Efficient Parallel Extraction of Crack-Free Isosurfaces from Adaptive Mesh Refinement (AMR) Data
This Week’s Computing Sciences Seminars
Absolute Value Preconditioning for Symmetric Linear Systems and Eigenvalue Problems
Monday, October 8, 11:00 am–12:00 pm, 50F-1647
Eugene Vecharynski, Georgia State University
Construction of the efficient preconditioning schemes for linear systems and eigenvalue problems represents one of the major challenges in scientific computing. Traditionally, a preconditioner has been defined and constructed as an approximation of the inverse of the coefficient matrix. In this talk, however, we show that for symmetric indefinite linear systems and interior eigenvalue problems it can be favorable to construct (symmetric positive definite) preconditioners which, instead, resemble the inverse of the absolute value of the (shifted) coefficient matrix. We call this strategy the “absolute value preconditioning”.
We present a practical example of the preconditioner based on the suggested approach. In this example, we construct the multigrid absolute value preconditioner which is applied for solving real Helmholtz type problems and accelerating the computation of a few interior eigenpairs of the discrete Laplace operator. The effectiveness of the new preconditioning strategy is demonstrated in our numerical tests.
Par Lab Seminar: Cloud Robotics
Tuesday, October 9, 1:00–2:30 pm, 430 Soda Hall (Wozniak Lounge), UC Berkeley
Damon Kohler, Google
A world filled with personal robots is inevitable but a lack of strong software infrastructure to support those robots is slowing progress. By offloading CPU and data intensive computation to the cloud, we can make robots lighter, cheaper, and smarter. The same developers that work on commodity web and mobile apps today can accelerate the pace of robotics research and development if we make high functioning robots affordable and universally accessible.
EECS Colloquium: The Road Ahead for Wireless Technology: Dreams and Challenges
Wednesday, October 10, 4:00–5:00 pm, 306 Soda Hall (HP Auditorium), UC Berkeley
Andrea Goldsmith, Stanford University
Wireless technology has enormous potential to change the way we live, work, and play. Future wireless networks will support Gigabit per second multimedia communication between people and devices with high reliability and uniform coverage indoors and out. Software will create a virtual wireless network cloud, enabling resource management, seamless connectivity, and roaming across heterogeneous access networks, including WiFi and cellular systems. Wireless technology will also enable smart and energy-efficient homes and buildings, automated highways and skyways, and in-body networks for analysis and treatment of medical conditions. The shortage of spectrum will be alleviated by advances in cognitive radios, and breakthrough energy-efficiency algorithms and hardware will be employed to make wireless systems “green”. There are many technical challenges that must be overcome in order to make this vision a reality. This talk will describe what the wireless future might look like and some of the innovations and breakthroughs that are required to realize this vision.
Advanced Graphics and Visualization Techniques with MATLAB
Thursday, October 11, 1:30–3:30 pm, 50 Auditorium
Saket Kharsikar, MathWorks Application Engineer
This session will focus on visualizing data, viewing images, and manipulating graphics in MATLAB. We will explore techniques for customizing graphical displays, generating animations, and creating publication quality graphics. We will present approaches to working with and displaying large data sets and images, and will discuss data importing, block-processing and re-sampling. Finally, we will investigate the visualization of higher-dimension data, with a focus on volumetric slicing and vector fields.
- Introduction to Handle Graphics
- Creating Animation
- Customizing Graphics
- Analysis of Large Data and Imagery
- Techniques for Higher- Dimension Visualization Parallel Computing with MATLAB
Register at http://www.mathworks.com/seminars/LBNL12 (although this is not mandatory, it will help us plan for the event).
Link of the Week: Sexists in White Coats: Men Favored for Laboratory Jobs
Decades into the post-feminist era, there are still pockets of society where women are held back from advancement due to pervasive stereotypes. If that reminder conjures up images of a military base or a corporate boardroom, think again. We’re talking about university science laboratories.
“Both male and female faculty judged a female student to be less competent and less worthy of being hired than an identical male student,” a Yale University research team reports in a disturbing new study appearing in the Proceedings of the National Academy of Sciences. Specifically, the researchers found that a female applying for a job as a laboratory manager was offered “a smaller starting salary and less career mentoring” than a male applicant with identical qualifications. 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.