InTheLoop | 04.11.2011
April 18, 2011
Tapia Participants Visit Berkeley Lab; Several Apply for Positions
Approximately 20 participants in the Richard Tapia Celebration of Diversity in Computing Conference visited Berkeley Lab on Wednesday, April 6, to learn more about careers in computing sciences and computing research at Berkeley Lab. Tapia 2011 was held at the Fairmont Hotel in San Francisco from April 3-5.
Jonathan Carter, Computing Sciences Deputy, welcomed the students and gave an overview of the facilities and research projects that make up the Computing Sciences Directorate at Berkeley Lab.
Marcus Day of CRD’s Center for Computational Sciences and Engineering talked about his efforts to computationally dissect combustion flames and how his findings are applied to experiments in the combustion laboratory of Robert Cheng of the Environmental Energy Technologies Division.
Greg Bell of ESnet spoke about the challenges of running a network optimized for international scientific collaborations and large datasets. The day culminated with tours of the Advanced Light Source, EETD’s Combustion Lab, and a talk about cyber forensics by Sean Peisert of CRD’s Complex Systems Group.
Organizers and hosts of the April 6 trip to Berkeley Lab include Computing Sciences’ Linda Vu, Jon Bashor, Yeen Mankin, Margie Wylie, and Jeff Todd.
The tour and the Lab’s participation in the Tapia conference itself are already yielding concrete results. Juan Meza, Acting Director of the Computational Research Division, reports that several participants have requested summer internships or postdoc positions.
ANI Testbed Departs Left Coast for Long Island
In another milestone, ESnet’s testbed was dispatched to its more permanent home at Brookhaven National Laboratory. The testbed, part of the $62 million ARRA-funded Advanced Networking Initiative (ANI), was established so researchers could experiment and push the limits of network technologies. A number of researchers have taken advantage of its capabilities so far, and ESnet is collecting proposals for new projects. Read more.
Google Offers 1 Billion Core-Hours to Researchers
Former Computational Research Division researcher David Konerding, now a Senior Software Engineer at Google and Advisor at Google Ventures, was instrumental in starting a new academic research grant program called Google Exacycle for Visiting Faculty, which was inspired in part by DOE’s INCITE program. Through this program, Google will award up to ten qualified researchers with at least 100 million core-hours each, for a total of 1 billion core-hours.
The program is focused on large-scale, CPU-bound batch computations in research areas such as biomedicine, energy, finance, entertainment, and agriculture, amongst others. For example, projects developing large-scale genomic search and alignment, massively scaled Monte Carlo simulations, and sky survey image analysis could be an ideal fit. Berkeley Lab researchers are encouraged to apply; the deadline is May 31, 2011. Read more.
Mentoring Program Deadline Next Week, Info Session Tomorrow
The CS 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 or Deb Agarwal. If you would like to learn more about the program, an information session will be held tomorrow, April 12, from 11 am to 12 pm in 50B-4205.
ESnet’s Inder Monga Demystifies Bits and Bytes
The latest entry in Berkeley Lab’s Video Glossary features ESnet network engineer Inder Monga explaining the difference between bits and bytes and how those terms are used in various contexts. Watch video.
NERSC’s Jason Hick Presents I/O Requirements for Exascale
The insideHPC website has posted a video of NERSC Storage Systems Group Lead Jason Hick’s presentation “I/O Requirements for Exascale,” recorded at the 2011 Open Fabrics International Workshop in Monterey. Slides from the workshop are now available for download, but you’ll need to complete a free registration with OFED first.
This Week’s Computing Sciences Seminars
Par Lab Seminar: Executable Biology: Successes and Challenges
Tuesday, April 12, 12:30–2:00 pm, 521 Cory Hall (Hogan Room), UC Berkeley
Jasmin Fisher, Microsoft Research, Cambridge, UK
As time goes by, it becomes more and more apparent that the puzzles of life involve more and more molecular pieces that fit together in increasingly complex ways. Biology is not an exact science. It is messy and noisy, and most often vague and ambiguous. We cannot assign clear rules to the way cells behave and interact with one another. And we often cannot quantify the exact amounts of molecules, such as genes and proteins, in the resolution of a single cell. To make matters worse (so to speak), the combinatorial complexity observed in biological networks (e.g., metabolic and signalling pathways) is staggering, which renders the comprehension and analysis of such systems a major challenge. Recent efforts to create executable models of complex biological phenomena — an approach called Executable Biology — entail great promise for new scientific discoveries, shading new light on the puzzle of life. At the same time, this “new wave of the future” forces Computer Science to stretch far and beyond, and in ways never considered before, in order to deal with the enormous complexity observed in biology. In this talk, I will summarize some of the success stories in using formal methods to model biology, as well as some of the major milestones on the way to conquer biological complexity.
LAPACK Seminar: Sparse Matrix Techniques in a Parallel Hybrid Solver for Large-Scale Linear Systems
Wednesday, April 13, 11:10 am–12:00 pm, 380 Soda Hall, UC Berkeley
Ichitaro Yamazaki, LBNL/CRD
A parallel hybrid (direct/iterative) linear solver based on the Schur complement method has a great potential to utilize thousands of processors for solving large-scale linear systems that are becoming increasingly difficult to solve using standard techniques. In this talk, we outline the algorithm implemented in our parallel hybrid solver, and discuss the sparse matrix techniques used for achieving high-performance. We also present numerical results of solving highly indefinite linear systems from real applications.
TRUST Security Seminar: Emerging Cyberspace R&D Themes Toward Trustworthy Systems
Thursday, April 14, 1:00–2:00 pm, 540a/b Cory Hall, UC Berkeley
Fredrick Sheldon, Oak Ridge National Laboratory
Cybersecurity, which comprises numerous interrelated components, and software assurance are inextricably intermingled. The former extends the boundary of physical security to the domain of cyberspace, while the latter provides the means for delivering on the promise that we can depend on the technologies that implement cyberspace. Secure systems must be dependable, and dependable systems fail if they are not secure. Unreliable software is inherently insecure. Unfortunately, cybersecurity practice and policy are largely heuristic, reactive, and increasingly cumbersome, struggling to keep pace with rapidly evolving threats. Advancing beyond this predominantly reactive posture will require a transformation in computing and communication systems architectures. New capabilities are required that don’t merely solve today’s plethora of security enigmas, but enable comprehensive game-changing strategies such as:
- Moving target, systems that move in multiple dimensions to the attacker’s disadvantage and to increase resiliency.
- Tailored trustworthy spaces, security tailored to the needs of a particular transaction rather than the reverse.
- Cybereconomic incentives, a landscape of incentives that reward good cybersecurity and ensure that crime doesn’t pay.
Fast Fourier Transform: Why? How? and the Course 842
Friday, April 15, 1:00–2:00 pm, 299 Cory Hall, UC Berkeley
Shamgar Gurevich, University of Wisconsin
The Discrete Fourier Transform (DFT) is one of the most important operators in computational mathematics. The DFT operator acts on the n-dimensional Hilbert Space L2(Z/n) of complex valued functions on the group of integers modulo n. It becomes very useful in the last century due to the Cooley-Tukey Fast Fourier Transform (FFT) algorithm that computes the DFT in order of n·/log(n) operations. In the lecture I will elaborate on an idea—due to Auslander and Tolimieri—which establishes this algorithm as a logical consequence of the Stone-von Neumann theorem in the symmetry theory (known also as representation theory) of the Heisenberg group. It is the hope of the lecturer that this point of view will leads to other FFT-like algorithms in signal processing.
The target audience for the lecture are undergrad/graduate students and post-docs, as well as researchers from EE/CS/Math. The main goal is to show how pure mathematical structures appear naturally in signal processing, and to help to promote the interactions between the EE/CS/Math communities. The lecture is based on the special applied algebra course (known as course 842) that I give this semester for EE/CS/Math students at UW Madison.
Link of the Week: Watson Takes a Turn on Wall Street
In the wake of Watson’s dominant performance on Jeopardy! last month, IBM has taken the technology on the road to showcase it to anyone who’ll listen. The supercomputer’s next task will be to apply its analytic smarts to healthcare applications, and IBM is pitching it to the financial markets too. An HPCwire article reports on these developments and also describes in detail how IBM engineers boosted Watson’s processing speed until it was fast enough to compete on Jeopardy. 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.