InTheLoop | 11.24.2008
The weekly newsletter for Berkeley Lab Computing Sciences
November 24, 2008
LBNL Team Wins Special ACM Gordon Bell Prize for Algorithm Innovation
A team of scientists from the Computational Research Division, led by Lin-Wang Wang, has won a prestigious Gordon Bell Prize, sponsored by the Association for Computing Machinery (ACM), for special achievement in high performance computing for their research into the energy harnessing potential of nanostructures. Their method, which was used to predict the efficiency of a new solar cell material, achieved impressive performance and scalability.
The ACM Gordon Bell Prize annually recognizes the best performance of scientific applications on supercomputers. This year’s prize, presented in a special category for algorithm innovation, was announced Thursday, Nov. 20, at the awards session of the SC08 conference in Austin.
ESnet and Collaborators to Develop and Test 100 GbE Technologies
Internet2, ESnet, Infinera, Juniper Networks and Level 3 Communications announced at the annual SC08 conference they have agreed to work together to aggressively develop and test emerging 100 Gigabit Ethernet (GbE) technologies. The initiative will build on fresh-from-the-laboratory technologies to create a 100 GbE testbed on the Internet2 and ESnet networks with an operational network capability soon thereafter. Read the complete story at http://www.lbl.gov/CS/Archive/news111808.html.
Lab Is Sponsoring Fall Cleanup This Week
The Laboratory is sponsoring a Fall Cleanup to remove unwanted materials at no cost to requesters. Facilities will accept requests through November 30, and they will be handled on a first-come, first-served basis. Requests are expected to be completed by December 5. Material must be under 500 lbs, and no hazardous waste or universal e-waste will be accepted.
Request a pickup online at Work Request and click the “cleanup” box. Include a description of all items and attach a Transportation Authorization Form (TAF) to all items.
For information about hazardous or universal e-waste pickups, contact Safety Coordinator Betsy MacGowan or Building Managers John Hutchings or Bill Iles.
Dark Energy Is Topic for Science at the Theater
The revolutionary discovery that the expansion of the universe is speeding up, not slowing down from gravity, means that 75% of our universe consists of mysterious dark energy. Berkeley Lab has played a key role in this research, traced from why the dinosaurs became extinct to the supernova breakthrough a decade ago to design of a new space telescope to reveal the nature of the unknown physics ruling our universe.
“Dark Energy Rules the Universe (and Why the Dinosaurs Don’t)” will be the topic for Berkeley Lab Friends of Science’s Science at the Theater presentation from 5:30 to 7:00 this evening (November 24) at Berkeley Rep’s Thrust Stage, 2025 Addison Street in downtown Berkeley. The speaker will be Eric Linder, Director of the Institute for Nuclear and Particle Astrophysics at Berkeley Lab and Deputy Director of the Berkeley Institute for Cosmological Physics. Admission is free. Here is the abstract:
Although ten years has passed since its discovery, the nature of dark energy is still unknown and has been termed the most important problem facing 21st century physics.
Join Dr. Eric Linder as we delve into the mystery of Dark Energy ten years later.
This Week’s Seminar Schedule
Monday, Nov. 24, 4–5 p.m., Soda Hall, Wozniak Lounge, UC Berkeley
Iterative Methods in Combinatorial Optimization
Mohit Singh, Microsoft Research, New England
In this talk we will demonstrate iterative methods as a general technique to analyze linear programming formulations of combinatorial optimization problems. We will focus on degree bounded network design problems where the task is to minimize the cost of the network and also satisfy given degree bounds on nodes. The most studied problem in this class is the Minimum Bounded Degree Spanning Tree problem. We present a polynomial time algorithm that returns a spanning tree of optimal cost while exceeding the degree bound of any vertex by at most an additive one. This is the best possible result for this problem and settles a 15-year-old conjecture of Goemans affirmatively. We will also discuss extensions to degree constrained versions of more general network design problems and give first additive approximation algorithms using the iterative method. These results add to a rather small list of combinatorial optimization problems which have an additive approximation algorithm.
This talk will contain joint works with Lap Chi Lau, Seffi Naor, Mohammad Salavatipour and R. Ravi. Tuesday, Nov. 25, 4–5 p.m., 258 Cory Hall, UC Berkeley
Fuzzy Communication by Context Dependent Reconstruction
Laszlo Koczy, Szechenyi Istvan University (Gyor) and Budapest University of Technology and Economics, Hungary
Fuzzy communication has been the subject of research interest for several decades now. Communication among humans often includes very effective contractions and omits all unnecessary parts, this way reducing the amount of communication elements (words, gestures, etc.), and time. Such reductions are often based on fuzzy elements in the message itself, and also in the common knowledge and mutually understood background information that allow the reconstruction of the full information intended to be sent by the communicating parties.
When overviewing the topic of “fuzzy communication” two antagonistically different interpretations may be found in the literature. One is discussing the possibility of deploying such fuzzy communication elements in intelligent man-machine and machine-machine “conversations”, in order to achieve more effective and flexible solutions, such as the research project run in the 1990’s at the Laboratory of Fuzzy Engineering Research in Yokohama, lead by Terano, or the more recent considerations on the use of fuzzy code books at the end points by Pedrycz. These researches aim at the creation of widely applicable and highly effective new technologies using reduced resources. The other part is considering fuzzy communication as a phenomenon in business and management which often leads to disastrous misunderstandings and consequently to losses and other disadvantages in company management, such as the work of Flaherty.
In this talk we want to point out that fuzzy communication can be used efficiently in practice if the proper algorithmic and software technologies are provided. After a short overview of previous projects we give some examples for human-human communication and then we revisit the cooperating and communicating intelligent robots project at LIFE while proposing some new aspects and fuzzy tools for realizing a multi-robot system based on intention guessing by context dependent information reconstruction. The proposed new approach uses fuzzy decision trees and fuzzy signatures for describing both the background code book and the actual situations and behavior of participating robots.
Reference will be given to several ongoing research grant projects where cooperating intelligent and manufacturing robot systems will be constructed by utilizing the above ideas. Tuesday, Nov. 25, 4–5 p.m., 540 Cory Hall, UC Berkeley
CHESS Seminar: Process Network in Silicon: A High-Productivity, Scalable Platform for High-Performance Embedded Computing
Mike Butts, Ambric, Inc., Beaverton, Oregon
Conventional high-performance embedded system technology has reached fundamental scaling limits of single CPU/DSP performance and ASIC/FPGA hardware development productivity. Adapting the SMP multicore programming model from general-purpose computing has serious scaling and reliability issues as well, as shown by work here at Berkeley. Adapting the SIMD programming model from HPC is not always a good match to increasingly irregular and complex algorithms in video, image and wireless processing.
We defined a Structured Object Programming Model, based on the Kahn process network model of computation, specifically for embedded computing. Objects are strictly encapsulated software programs running concurrently on a MIMD array of processors and memories. They communicate and synchronize with one another through a structure of buffered channels, in which every data transfer is a synchronization event. An application is a hierarchy of objects and structure, programmed directly in high-level language, without need for automatic parallelizing compilers or synthesis tools.
Having first defined this programming model, we defined a scalable hardware architecture, developed one teraOPS production silicon and an integrated development environment, and delivered it to a number of customers who are designing it into their products. University projects have been using it as well. Developers have found this programming model easy to learn and use, and two to three times as productive as their previous DSP, FPGA or ASIC platforms.
A hardware and tools demo will be available after the seminar.
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 7,000-plus 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 Department of Energy 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.