A-Z Index | Phone Book | Careers

InTheLoop | 10.18.2010

October 18, 2010

Web Page Highlights Computational Environmental Research

Environmental researchers rely on computers to expose the fundamental workings of our planet and atmosphere, to bring complicated questions to light and find solutions. By developing computational tools to effectively monitor Earth’s evolution and to model climate and new technologies, scientists in Berkeley Lab’s Computational Research, Earth Sciences, and Environmental Energy Technologies divisions are ensuring that society will successfully adapt to a changing world. Read more.

HPCwire Cites Green Flash in Essay on Specialized Supercomputers

Although the Green Flash research project has a lower profile these days than in 2008 when it was introduced, it still hasn’t been forgotten. In the October 14 article “For Proprietary HPC, Hope Springs Eternal,” HPCwire editor Michael Feldman surveys the landscape of specialized HPC systems and includes a paragraph on Green Flash, the specialized climate modeling system under development at Berkeley Lab.

Although the article equates specialized with proprietary, a comment posted at the end of the article points out that Green Flash is both specialized and based on a commodity configurable architecture—a combination common in the consumer electronics market but new to the HPC domain.

Berkeley Lab Staff Contributing to IEEE VisWeek Next Week

IEEE VisWeek 2010 will be held in Salt Lake City, Utah on October 24–29, and members of CRD Visualization Group and NERSC Analytics Team will be contributing to a tutorial and a paper. Hank Childs co-organized the tutorial on “Large Vector-Field Visualization: Theory and Practice.” Gunther Weber co-authored a paper on “Two-Stage Framework for a Topology-Based Projection and Visualization of Classified Document Collections.” Also, Inna Dubchak of Berkeley Lab’s Genomics Division will participate in a panel discussion of “Challenges in Visualizing Biological Data.”

Acclaimed Play A Disappearing Number to Be Broadcast to Theaters

Robert L. Bryant, Director of the Mathematical Sciences Research Institute (MSRI), has announced to friends of MSRI an opportunity to see a British production of the play A Disappearing Number broadcast to local theaters. The play opened in Plymouth in 2007 and received several awards in the UK for best new play.

Described as “a compelling meditation on love, mathematics and the pain of exile in an age when we think we can belong anywhere and have everything,” A Disappearing Number weaves together the story of two love affairs, separated by a century and a continent. It tells of the heartbreaking collaboration between the greatest natural mathematician of the 20th century, Srinivasa Ramanujan, a penniless Brahmin from Madras in South India, and his British counterpart, the brilliant Cambridge don G.H. Hardy.

A Disappearing Number was broadcast by National Theatre Live from the Theatre Royal Plymouth on October 14 and will be screened in these local theaters:

In the East Bay:

  • Rialto Cinemas Elmwood — Tel: 510 433 9730
    19th & 21st October 2010
  • Rialto Cinemas Cerrito — Tel: 510 273 9102
    1st & 3rd November

In San Francisco:

  • Sundance Kabuki — Tel: 415 929-4650
    2nd & 6th November 2010

This Week’s Computing Sciences Seminars

The 3-D Revolution: Scientific Visualization in Three Dimensions
Tuesday, October 19, 9:30–10:30 am, 50F-1647
Joerg Meyer, University of California, Irvine

The recently announced commitment of film production companies to release many of their future movies in 3-D has spurred a renewed interest among the public and among researchers in the area of three-dimensional imaging and scientific visualization.

Technologies that were developed decades ago in research and science are now becoming mainstream. As it frequently happened in the past, the gaming industry is taking the lead and is pushing forward cutting-edge technology for a growing consumer market.

This presentation will give an overview of recent technological improvements in the field of three-dimensional rendering primarily for scientific visualization applications. We will discuss how the availability of such enabling technologies changes the face of modern data analysis and visual analytics.

LAPACK Seminar: Reduced Rank Regression via Convex Optimization
Wednesday, October 20, 11:10 am–12:00 pm, 380 Soda Hall, UC Berkeley
Ming Gu, UC Berkeley and LBNL/CRD

Reduced rank regression is a well-known technique for dimension reduction and coefficient estimation for multivariate linear regression. In this talk, we discuss the formulation of various reduced rank regression models, using convex optimization techniques; we also develop a general solution technique, based on spectral projection, to efficiently solve such problems. We present numerical experiments that demonstrate the effectiveness of our method.

Visualization and Analysis of Multi-Dimensional Scientific Data
Wednesday, October 20, 11:30 am–12:30 pm, 50B-2222
Oliver Ruebel, Lawrence Livermore National Laboratory

Knowledge discovery from large and complex scientific data is a challenging task. Modern simulation and data acquisition methods allow simulation and measurement of natural and man-made phenomena at unprecedented levels of detail and play a key role in advancing scientific knowledge in grand challenge research areas such as developmental biology, high-energy physics, climate modeling and design of new clean energy production methods and complex machinery. The growing numbers of data dimensions and data objects (such as particles) in modern scientific data result in tremendous challenges for visualization, data analysis, and data exploration methods and algorithms. Researchers are overwhelmed with data, and new approaches are needed to enable effective data analysis and knowledge discovery.

To address these challenges we develop, combine, and integrate methods from scientific visualization, information visualization, automated data analysis, and other enabling technologies, such as efficient data management. This integrated approach has proven to be effective in a diverse set of application areas ranging from developmental biology to high-energy physics. The seminar will focus in particular on visualization and analysis of data from numerical simulations of laser wakefield particle accelerators (LWFAs). In this context, we will describe novel methods for: (i) fast, query-based visual exploration of extremely large particle data and (ii) efficient, automatic detection and classification of particle beams in LWFA simulation data. The close integration of the visualization system VisIt with state-of-the-art data management using FastBit allows us to quickly identify and trace particles of interest, enabling a more efficient and accurate visual analysis than previously possible. Manual exploration, in particular of a large number of simulation data sets, is time-consuming. To address this challenge, we developed novel algorithms for automatic detection of particle beams based on a step-by-step analysis process. In each step we derive additional information about the particle beams while at the same time reducing the amount of data we need to consider in the subsequent analysis. This approach, in combination with advanced data management using FastBit, supports efficient and accurate classification of particle beams based on the complete history of the particles that form them.

TRUST Security Seminar: Return-Oriented Programming: The Impact of the Gadget on Civilization
Thursday, October 21, 1:00–2:00 pm, Soda Hall, Wozniak Lounge, UC Berkeley
Hovav Shacham, University of California, San Diego

Return-oriented programming is an attack technique that induces arbitrary behavior in the compromised program without injecting new code into its address space. A return-oriented attack combines short sequences of instructions from a target program's executable image into a Turing-complete set of combinators, called “gadgets,” from which any desired functionality can be synthesized. The addresses of the desired sequences are arranged on the program’s stack, which the attacker rewrites using a buffer overflow or other memory vulnerability. Each instruction sequence used ends in a “return” instruction; this instruction transfers control from the currently executing sequence to the next (and gives the technique its name). Return-oriented programming was introduced in 2007 as a way of defeating the Data Execution Prevention defense that Microsoft implemented in Windows XP SP2. The original attack was specific to the x86 processor and largely manual.

In this talk, we survey the research on return-oriented programming since its introduction, showing that the threat is both more general and more acute. Defenses that prevent code injection (without also protecting control flow integrity) are useless against return-oriented programming. In particular, relative to return-oriented programming, Harvard architecture machines (with separate data and program memories) are no more secure than von Neumann architecture machines (with one data-and-program memory). Return-oriented programming is applicable across a range of architectures and systems: research has demonstrated it on the x86, SPARC, Z80, PowerPC, and AVR processors; on Linux, Solaris, Windows, and embedded systems; on PCs, servers, routers, sensor network nodes, and even voting machines. Automated tools like exploit compilers and gadget-set generators can make writing return-oriented exploits as simple as writing traditional ones.

Resistive Random Access Memory (RRAM) for Next-Generation Nonvolatile Memory Applications
Friday, October 22, 1:00–2:00 pm, 521 Cory Hall (Hogan Room), UC Berkeley
An Chen, Strategic Technology Group, Globalfoundries, Sunnyvale, CA

With Flash memory approaching the CMOS scaling limit, various novel nonvolatile memory concepts have emerged. Among them, resistive switching memories, also known as RRAM for Resistive Random Access Memory, have shown promising characteristics. Although several emerging memories are based on resistive elements (e.g., phase change memory, MRAM, etc.), RRAM usually refers specifically to metal-oxide based resistive switching devices. These devices are built in a metal-insulator-metal structure and can be electrically switched between two or more stable resistance levels. Numerous metal oxides have demonstrated resistive switching behaviors, including NiOx, TiOx, HfOx, CuOx, ZrOx, SrTiO3, PrxCa1-xMnO3, etc. These devices can be switched with low power and fast speed, have demonstrated long retention and good endurance, and are believed to be highly scalable and stackable. Although the switching mechanisms are not completely understood, they are often attributed to some defects such as mobile ions, charge traps, oxygen vacancies, etc. This presentation will provide an overview of the switching mechanisms and electrical characteristics of RRAM. Opportunities and challenges of RRAM devices will be discussed. CuOx-based RRAM devices will be presented as a detailed example of industry emerging memory research.

Programming Language Memory Models: What Do Shared Variables Mean?
Friday, October 22, 3:00–4:00 pm, 540 Cory Hall, UC Berkeley
Hans Boehm, HP Labs

Although multithreaded programming languages are common, there has been a surprising amount of confusion surrounding the basic meaning of shared variables. There is finally a growing consensus, both that programming languages should by default guarantee an interleaving-based semantics for data-race-free programs, and on what that should mean. We discuss the motivation for, and both implementation and user consequences of, this guarantee.

Unfortunately, it is also increasingly clear that such a guarantee, though useful, is insufficient for languages intended to support sand-boxed execution of untrusted code. The existing solution in Java is only partially satisfactory. The solution to this problem is far less clear. We briefly outline one promising approach.

Link of the Week: The Best Fiscal Stimulus: Trust

The potent hormone of empathy, oxytocin, is shaking up the field of economics, writes Michael Haederle in Miller-McCune Magazine. The neuroeconomist Paul Zak and his collaborators at Claremont Graduate University have found that oxytocin, a hormone produced in the brain that promotes human bonding, plays a powerful role in shaping how generous people are. Economic growth typically lags in countries with low levels of trust, because a business transaction needs to satisfy both parties, and that calls for trust. This new approach promises to shift economics away from a physics-based model to a focus on biological systems. 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.