InTheLoop | 10.06.2008
The weekly newsletter for Berkeley Lab Computing Sciences
October 6, 2008
Berkeley Lab Staff Participating in CS/Math Workshop This Week
A DOE workshop on CS/Math Institutes and High Risk/High Payoff Technologies for Applications will be held in Chicago this Tuesday–Thursday, October 7–9. It will run in two parallel tracks.
The High Risk technologies track will have presentations by several application developers describing the potential high risk/high payoff areas in their application domain. These will be followed by discussion by the attendees on how the math and CS communities could help make this high payoff a reality. For example, it is high risk to rewrite an application in one of the new HPCS languages but the payoff could be high down the road. Other ideas could include radical algorithms, or new solution approaches. The outcome of this track will be a list of potential high risk endeavors and a few illustrative examples. Phil Colella, leader of CRD’s Applied Numerical Algorithms Group (ANAG), will chair this track.
The concept of joint math and computer science institutes in a new one. The Institutes track will help refine the community's thoughts on how to create, develop, and manage joint CS/Math Institutes. There will be four presentations in this track with most of the time spent in discussion and refinement of the Institute concept. (Brian van Straalen of ANAG will give a presentation on adaptive mesh refinement.) Each Institute would involve both CS and math researchers working under some unifying theme with a mix of Lab and university participants. A couple examples of themes are math libraries, and programming models focused on the challenges of petascale computing. Other themes are expected to emerge from the workshop.
Other Computing Sciences staff participating in the workshop will be David Bailey, John Bell, Erik Roman, John Shalf, and Kathy Yelick.
A report detailing all the information and ideas that come out of the workshop will be posted at Workshop on CS/Math Institutes and High Risk / High Payoff Technologies for Applications.
BER Associate Director Anna Palmisano Visits CS and NERSC
Dr. Anna Palmisano, Associate Director of the Office of Biological and Environmental Research (BER) in the DOE Office of Science, visited Computing Sciences last week. ALD Horst Simon discussed computational research collaborations and ESnet with her, and NERSC Director Kathy Yelick gave her a tour of NERSC.
In a thank-you note, Dr. Palmisano wrote, “I have heard so much about the Center over the years I have been associated with BER, so it was quite a thrill to actually have a chance to see it in action. NERSC is playing a vital role in many of our BER programs, and I expect that will continue and grow in areas such as climate and computational biology.”
Cyber Security Update: Good News and Ongoing Challenges
IT Division Director Rosio Alvarez sent this memo on September 30:
First, the Lab experienced a slightly higher number of incidents, but the total damage from this was the lowest since 2001. A thank you goes to you and your staff for appropriately securing your systems.
Our major incidents this year can be grouped into three categories:
1. Ongoing issues related to stolen credentials (usernames and passwords)
2. Increasingly sophisticated phishing and social engineering
3. Attacks on web servers.
You can read more about each of these attacks on the CPP website at Berkeley Lab Cyber Security but I would draw your attention in particular to #2. The entire .edu and .gov space has seen a significant increase in highly targeted attacks aimed at specific affiliations, research areas, and university connections. This is an area where awareness is crucial, remember, always think before you click and be on the defensive.
There are a number of challenges ahead with regards to DOE and Federal policy as well. DOE has released, in near final form, thousands of pages of new cyber security requirements — many of which are inappropriate for an open research facility like Berkeley Lab. These include requirements that turn our posture from open to closed. I led the National Labs in developing a coordinated position against these Directives this year and, while the final versions are substantially improved, from our perspective they still contain many disruptive and expensive requirements. We will be engaging the Office of Science for a number of risk-based waivers and hope, in this way, to minimize the risk and disruption for you in conducting open science.
I am happy to discuss these issues with you and your divisions at any time. I want to thank you again for you and your staff’s work toward keeping our cyber environment secure.
Advance Registration for SC08 ends on October 15
What do orthodontics, clothes dryers, and fighter jets have in common? They are all being improved with high performance computing, communications and storage technologies. Find out more by claiming your spot at SC08. Register before Wednesday, October 15, to take advantage of the advance registration discounts.
SC08 will celebrate 20 years of unleashing the power of HPC this year, following the traditions set by the first SC Conference in 1988. The conference, which opens November 15 in Austin, showcases the latest in high performance computing, networking, storage, and analysis applied to the world’s most challenging commercial, scientific, and engineering problems.
Registration for SC08 Technical Program is now open. This year’s Technical Program will include two days of tutorials, three days of technical paper presentations, a keynote address and other invited speakers, seven panel discussions, a flock of birds-of-a-feather sessions, poster presentations and more. For more information about registration categories and rates and to register, go to: SC08 - Registration.
Promoting Safety Culture in Computing Sciences
In response to ALD Horst Simon’s recent email concerning safety culture, a Computing Sciences summer intern sent the following safety success story:
In response to “safety culture,” I just wanted to praise the attention re. safety I’ve had from both [my group lead and supervisor] since I started interning at the lab this summer. As it turns out, I had been developing wrist problems throughout the previous year, and I hadn’t made the connection yet that it was caused by some bad work habits from my CS coursework! [My group lead and supervisor] were both insistent about my getting an ergo evaluation within the first few days I started work, and helpful in setting up my workstation properly, which proved invaluable in eliminating my wrist problems this summer. Also, the lab’s online resources for ergonomics were especially useful. On another safety topic, some members of the Lab’s OCFO and EHS staff have made two short videos offering tips on how to reduce the potential for injury when moving to a new office. CS staff who may have to move during the Building 50 remodeling — or for any other reason — are encouraged to watch these videos:
This Week’s Seminar Schedule
Wednesday, Oct. 8, 11:10 a.m.–noon, 380 Soda Hall, UC Berkeley
Matrix Inversion in Nanoscale Device Simulation
Douglas Mason, Harvard University Physics
I will present a class of problems from condensed matter physics which map onto numerical matrix inversion. These problems will be key to developing the next class of transistors for the 21st century and have earned a large amount of attention in physics journals. Since physicists aim to scale these problems to larger-and-larger systems, there has been great incentive to optimize their matrix inversion routines. Additional restraints, and limitations on the information desired, have spawned a number of approaches which can adjust themselves to the problem at hand for maximum efficiency. I will present two state-of-the-art methods on the verge of distribution being fashioned at LBNL and Stanford. In addition, I will demonstrate the LBNL method on some sample systems of great interest to the nanotechnology community, that is, disordered graphene nanoribbons. Wednesday, Oct. 8, noon–1:00 p.m., 290 Hearst Memorial Mining Bldg., UC Berkeley
CITRIS Research Exchange: The Future of Service Science
Paul Maglio, Adjunct Associate Professor of Cognitive Science, UC Merced, and Senior Manager of Service Systems Research at IBM Almaden Research Center
Live online broadcast: mms://media.citris.berkeley.edu/webcast
Service Science is the study of service systems — configurations of people, technologies, and other resources that interact with others to create mutual value — and it aims to develop theory and practice around service innovation. Service Science has come a long way in a short time. Just a few years ago, no one had heard of it, and now there are new courses, programs, degrees, and even departments related to service science popping up all around the world. How did we get here? In this talk, I will trace the (short) history of service science, discussing its historical context, its current motivations, and its prospects for becoming a discipline of its own. Thursday, Oct. 9, noon–1:00 p.m., Sibley Auditorium, Bechtel Engineering Center, UC Berkeley
View from the Top Lecture Series: Shaping the Future with Technology
Craig Mundie, Chief Research and Strategy Officer, Microsoft
Please join us for a special presentation by Craig Mundie, chief research and strategy officer of Microsoft. Craig is one of two senior executives who took over from Bill Gates and is responsible for directing the company’s long-term technical strategy and investments.
Craig will talk about the technologies that will change how people interact with the world, and the opportunities for every academic discipline to harness the power of computing. He will also showcase a number of advanced technologies, and answer questions from the audience. Thursday, Oct. 9, 4:00–5:00 p.m., 290 Hearst Memorial Mining Bldg., UC Berkeley
CITRIS Distinguished Speaker: The Brain-Like Vision
Edgar Koerner, President of the Honda Research Institute Europe GmbH
Live online broadcast: mms://media.citris.berkeley.edu/webcast\
Intelligence is a technology and a strategy for robust and flexible problem solving in complex environments (both natural and artificial) under the constraints of limited resources (e.g. time, energy). Still the brain is the only intelligent system of that kind we know of. Understanding essential principles of how the brain controls behavior may enable us to provide our technical artifacts at least with some aspects of its performance we admire. Our approach is based on the assumption that the essence of brain computing is not in the local processing or learning algorithm but in the way the brain organizes processing. The challenge of that approach is, first, that the brain is an inhomogeneous network of a huge number of local processors at several interacting levels of complexity.
There are highly specialized types of elementary processing elements, the neurons, which are in turn organized into static task-specific clusters which are organized within a macroscopic function-specific architecture; and all of which are subject of a behavior control that creates a dynamic clustering of processing resources across all complexity levels of systems organization. Second, any meaningful simulation of a large-scale hypothesis on brain function is hampered by the limitations enforced by the available technology which usually results in elimination of several levels of systems complexity via mean field concepts or other abstractions, thus collapsing the dimensionality of nested algorithmic structures of the models. As a consequence, we target brain processing control architectures at several levels of complexity in parallel: (1) We investigate the control of growth processes by gen-regulatory networks; (2) at the level of detailed cortical columnar architectures we are looking into self-referential control for storing experience; (3) the behavior based dynamic allocation of systems resources is targeted for visual scene analysis; and (4) the global behavior control architecture is explored in the context of step by step increasing capabilities for autonomous interaction of our humanoid robot ASIMO.
After substantiating the philosophy of our approach, the recent progress towards such a flexible cognitive control is shown for autonomous acquisition of visual information. For visual scene analysis a dynamical system of active processes in brain-inspired architecture is researched that can rapidly configure themselves under the control of static information both sensory and previously acquired, and then allow themselves to hand over control to changing sensory signals. That cluster of visual routines which resemble the function of the parietal visual pathway in primates provides with attention, fixation, tracking the necessary prerequisite that the subsystem modeled according to the ventral pathway can deal with a sequence of still frames in the focus of attention to acquire visual experience by on-line learning. This active vision system is being implemented into the behavior control system for autonomous interaction of our humanoid robot ASIMO. Step by step we implemented the nested control loops for reflexes, attention modulated behavior, on-line learning from sensory experience, and prediction/expectation driven behavior. The capability of ASIMO is demonstrated to learn recognize objects from interaction with human partners, and to learn associations between acoustic and visual objects, as well as the association of sound to behavioral concepts and eliciting simple prediction driven behavior. Finally, the same vision system implemented within a visually based safety system of a car shows reliable “braking to a halt” in a critical situation of a realistic road construction site.
Link of the Week: DARPA’s Math Quiz
On September 26, DARPA issued a request for proposals to solve 23 “mathematical challenges” such as developing a mathematical theory to build a functional model of the brain, solving the 150-year-old Riemann Hypothesis, and developing the high-dimensional mathematics needed to accurately model and predict behavior in biology and human interactions, among other ambitious goals. Wired Blog Network describes the challenges at DARPA Math Quiz. The DARPA solicitation can be found at DARPA Mathematical Challenge.
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.