A-Z Index | Phone Book | Careers

InTheLoop | 03.28.2011

March 28, 2011

Invisible Satellite Galaxies Revealed with NERSC Systems

Astronomers predict that large spiral galaxies, like our Milky Way, have hundreds of satellite galaxies orbiting around them. Using supercomputers at NERSC, researcher Sukanya Chakrabarti has developed a mathematical method to uncover these “dark” satellite galaxies. When she applied this method to our own Milky Way, Chakrabarti discovered that a faint satellite might be lurking on the opposite side of the galaxy from Earth. Read more.

DOE Workshop for Industry Software Developers on Thursday

The SciDAC Outreach Center, in collaboration with the DOE Office of Advanced Scientific Computing Research (ASCR) Facilities Division and Oak Ridge National Laboratory, is holding a DOE Workshop for Industry Software Developers this Thursday, March 31, in Chicago. The purposes of the workshop are to introduce commercial software developers and industry to SciDAC software and its developers; to explain the process to obtain this software at no cost; and to identify private sector software development requirements to help DOE better understand areas for collaboration and potential Small Business Innovation & Research (SBIR) topics.

NERSC’s David Skinner, head of the SciDAC Outreach Center, will present “SciDAC Program Overview”; Osni Marques of CRD will present “SciDAC Software Introduction”; and Tony Drummond of CRD will present “Case Study: Sundials/CVODE.”

Kevin Oberman Speaks on IPv6 at ASCAC Meeting

A March 21 InTheLoop item on the DOE Advanced Scientific Computing Advisory Committee (ASCAC) spring meeting held last week in Washington, DC, neglected to mention that ESnet network engineer Kevin Oberman spoke at the meeting on issues and implications of IPv6.

CS Mentoring/Networking Program Is Accepting Applications

The Computing Sciences Mentoring/Networking Program is now accepting applications, with a deadline of April 19. For the Mentor Enrollment Form, go here; for the Protégé Enrollment Form, go here. If you have any questions about the program, please contact Marcia Ocon Leimer.

Tapia Diversity in Computing Conference in S.F. Next Week

The Richard Tapia Celebration of Diversity in Computing Conference will be held at the Fairmont Hotel in San Francisco next week, April 3–5, 2011, chaired by David Patterson of CRD. Tony Drummond of CRD is leading the Student Research Poster Session and Dinner Reception. Elizabeth Bautista of NERSC will participate in a panel discussion of “Opportunities in High-Performance Computing at Department of Energy Laboratories.” Jon Bashor, Computing Sciences Communications and Outreach Manager, is serving as public relations chair.

Earthquake Safety Tips

How safe is your home in the event of a major earthquake? Is it true that you should stand in a doorway during an earthquake? What should you do if you are outdoors?

Those questions and many others are answered in Putting Down Roots in Earthquake Country, a 32-page earthquake science and preparedness handbook, updated yearly and available here in several languages and formats. Key steps to seismic safety are outlined in “Seven Steps to Earthquake Safety.”

Here are the answers to the second and third questions:

  • In the early days of California, many homes were made of adobe bricks with wooden doorframes. After a powerful earthquake, doorframes were sometimes only parts of these houses still standing. From this came the myth that a doorway is the safest place to be during an earthquake. Today, few people in the Bay Area live in old, unreinforced adobe houses. In modern houses, doorways may be no stronger than any other part of the house and do little to protect you from falling debris. You are safer under a table, so “DROP, COVER, AND HOLD ON.”
  • If you are outdoors, move to a clear area if you can safely do so; avoid buildings, power lines, trees, and other hazards. Always assume fallen power lines are live!

This Week’s Computing Sciences Seminars

Applications on Emerging Paradigms in Parallel Computing
Tuesday, March 29, 9:30–10:30 am, 50B-2222
and Thursday, March 31, 10:00–11:00 am, 50B-4205
Abhinav Sarje, Truveo, AOL Inc.

The area of computing is seeing parallelism increasingly being incorporated at various levels: from the lowest levels of vector/SIMD processing, SMT, and multi/many-cores with thread-level shared memory and SIMT parallelism, to the higher levels of distributed memory parallelism as in supercomputers and clusters, and scaling them to large distributed systems as server farms and clouds. All together these form a large hierarchy of computational and memory parallelism. Developing high-performance parallel algorithms and efficient software tools, which exploit the available parallelism, is inevitable in order to harness the raw computational power these emerging systems have to offer. In our work, we develop architecture-aware parallel techniques on such emerging paradigms in parallel computing, specifically, parallelism offered by the emerging multi- and many-core architectures, as well as the emerging area of cloud computing, to target large scientific applications.

Our work includes parallel algorithms to compute optimal genomic alignments on heterogeneous multi-core processors; and parallel architecture-aware techniques for scheduling all-pairs computations on heterogeneous multi- and many-core systems. We compare the performance of our strategies on Cell, graphics processors and Intel Nehalem processors. We apply our methods to specific applications taken from the areas of computational genomics, systems biology, fluid dynamics, and materials science.

In the area of cloud computing, we propose and develop an abstract framework to enable efficient parallel computations on large tree structures, to facilitate easy development of applications based on trees. Our framework, in the style of Google’s MapReduce paradigm, abstracts away the computational patterns from the applications to provide a simple and generic interface to a user to write their applications. We implement our framework as a generic programming library for a large cluster of homogeneous multi-core processors, and demonstrate its applicability through applications in scientific computing.

Par Lab Seminar: Enabling Parallelism for Productivity Programmers with Patterns, Frameworks, and Languages
Tuesday, March 29, 11 am–12:30 pm, Soda Hall, Wozniak Lounge, UC Berkeley
Ade Miller, Microsoft

Helping productivity programmers develop applications that run well on multicore hardware is one of the major challenges to broad adoption of parallel programming. This talk covers some of the work going on at Microsoft to enable Windows developers to write applications that target today’s multicore architectures. It gives an overview of the new frameworks, language features added in Visual Studio 2010 to support parallel programming and the patterns they enable.

Building Electronics from Solution
Wednesday, March 30, 12:00–1:00 pm, Sutardja Dai Hall, Banatao Aud., UC Berkeley
Live broadcast at mms://media.citris.berkeley.edu/webcast
Ana Arias, UC Berkeley

Solution-processed electronic materials have the potential to create a new manufacturing paradigm and applications domains beyond those now dominated by silicon technology. These materials can be deposited and patterned with tools commonly used in the graphics design and printing businesses. Over the past 10 years, solution-processed semiconducting materials have been studied largely for incremental application in information display. However, combining derivatives of these semiconductors with emerging solution-dispersible metal and metal oxide nanoparticles and nanowires enables the fabrication of electronic devices that are fully built from solution. This establishes a new device processing platform which, in turn, allows device form factors and integration of functionality in systems not feasible in any conventional semiconductor technology.

In this talk, I will discuss the development of a flexible integrated blast dosimeter to illustrate and demonstrate the challenges and advantages of using solution-processed electronic materials for flexible and disposable applications.

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.