September 7, 2010
insideHPC Names John Shalf a “Rock Star of HPC”
The online magazine insideHPC is doing a series of interviews with people they consider the “Rock Stars of HPC.” “Shalf’s energy and dedication to HPC are helping to actively shape the future of HPC, and that’s what makes him this month’s HPC Rock Star,” writes insideHPC’s Mike Bernhardt. The interview covers how Shalf got involved in supercomputing in his youth, highlights of his career, and his thoughts on the challenges facing HPC.
Shalf’s reaction to the “Rock Star” label? “No groupies.”
The “Rock Star” list so far includes Marc Snir, Associate Director for Extreme Scale Computing at the National Center for Supercomputing Applications (NCSA); Thomas Sterling, Professor of Computer Science at Louisiana State University; Ricky Kendall, leader of the Scientific Computing Group at the National Center for Computational Science at Oak Ridge National Laboratory; and Bill Kramer, former NERSC General Manager and now deputy project director and co-principal investigator for the Blue Waters project at NCSA.
Skinner to Talk at 4th Parallel Tools Workshop in Stuttgart
David Skinner of NERSC is giving a talk on IPM at the 4th Parallel Tools Workshop being held today and tomorrow (September 7-8) at the High Performance Computing Center Stuttgart (HLRS) of the University of Stuttgart, Germany.
Visualization Group Has Opening for Postdoctoral Fellow
The Visualization Group in CRD has an immediate opening for a Visualization Postdoctoral Researcher who will focus on the development of algorithms and their application to visualization, image analysis, and pattern recognition for nanoscale control of geologic CO2. Other relevant techniques include but are not limited to multivariate, temporal, large data, quantitative microscopy and statistical analysis. See job details. The Lab’s Employee Referral Incentive Program (ERIP) awards $1,000 (net) to employees whose referral of an external candidate leads to a successful hire.
SC10 Is Accepting Submissions for Video Loop
Over the years, SC has provided members of the HPC community the chance to showcase samples of their work in the video loops that ran before each of the plenary sessions. The resulting videos have been resounding successes, and this event will return to SC10, November 13–19, 2010 in New Orleans.
SC10 is accepting submissions for this year’s video loop now. Images can relate to any area of HPC, and will be an opportunity to show the quality and range of work that is ongoing within the community. Deadline for participation is Friday, October 15. Go here for more information on sharing your work at SC10. For questions or to request a participation form, get in touch at email@example.com.
This Week’s Computing Sciences Seminars
Solving Linear Systems of Equations with Aggregation-Based Multigrid: Theory and Applications
Wednesday, Sept. 8, 10:00–11:00 am, 50B-2222
Artem Napov, Service de Métrologie Nucléaire, Université Libre de Bruxelles, Belgium
Applications in science and engineering frequently require the solution of linear systems of equations. As the number of unknowns grows, driven, for instance, by the needs of better quality and higher accuracy of the modeling, the issues of solution time and computational resources become increasingly important. In such context, scalable iterative methods represent often an attractive (if not the only possible) approach. Among them, the family of multigrid methods is known to be particularly suitable for the solution of systems arising from discretization of partial differential equations.
In the present talk, multigrid methods based (unsmoothed) aggregation is discussed. A particular emphasis is placed on algebraic (or black-box) variants, which require only the information contained in the system matrix. Such methods are motivated by a relatively low setup cost and modest memory requirements. A proper implementation can also lead to a rapid solution. With the help of new theoretical results, we shed some light on the way an optimal implementation of the (algebraic) aggregation-based multigrid can be obtained. Numerical experiments demonstrating the potentialities of one such implementation are presented and discussed.
LAPACK Seminar: The History of “Gaussian” Elimination
Wednesday, Sept. 8, 11:10 am–12:00 pm, 380 Soda Hall, UC Berkeley
Gaussian elimination is universally known as “the” method for solving simultaneous linear equations. As Leonhard Euler remarked in 1771, it is “the most natural way” of proceeding. The method was invented in China about 2000 years ago, and then it was reinvented in Europe much more recently, so it is surprising that the primary European sources have not been identified until now. It is an interesting story in the history of computing and technology that Carl Friedrich Gauss came to be mistakenly identified as the inventor of Gaussian elimination even though he was not born until 1777. The European development has three phases. First came the “schoolbook” method that began with algebra lessons written by Isaac Newton; what we learn in high school or middle school is still basically Newton's creation. Second were methods that professional hand computers used to solve the normal equations of least squares problems; until comparatively recently the chief societal use for Guassian elimination was to solve normal equations for statistical estimation. Third was the adoption of matrix notation; henceforth the schoolbook lesson and the professional algorithms were understood to be trivially related in that all can be interpreted as computing triangular decompositions.
Local Codes: Mapping and Transforming a City’s Forgotten Public Space
Wednesday, Sept. 8, 12:00 pm, Banatao Auditorium, Sutardja Dai Hall, UC Berkeley
Live broadcast: mms://media.citris.berkeley.edu/webcast
Nicholas de Monchaux, UC Berkeley
Using the latest digital mapping technologies, “Local Code” documents the location and character of 1,500 City-owned “remnant parcels” in San Francisco. Conceived of by UC Berkeley professor Nicholas de Monchaux, the project uses geospatial analysis and parametric modeling to provide tools for citizens to transform these derelict pockets of public space into useful and beautiful places that, when combined, could contribute significantly to the City’s larger sustainability goals.
Silicon Photonics and the Data-Centric Datacenter
Friday, Sept. 10, 11:00 am–12:30 pm, 521 Cory Hall (Hogan Room), UC Berkeley
Marco Fiorentino, Hewlitt Packard
Recent trends show an explosion in the amount of data being created each year. In response to this trend new system architectures optimized for data-centric workloads are being developed. These architectures can only be brought to fruition by developing new technologies to transmit, compute, and store data. My talk will focus on the impact of photonics, and silicon photonics in particular, on these trends and discuss the work being done at HP labs to advance these technologies.
The Sensing Entity Tracking Initiative (SETI): A CITRIS Center for Smart Energy Infrastructure
Friday, Sept. 10, 12:00 pm, 250 Sutardja Dai Hall, UC Berkeley
Alberto Cerpa, UC Merced
SETI combines state-of-the-art network technology with novel data processing algorithms in order to determine how people are distributed within a building. The ultimate goal of the seed-funded project is to model and accurately predict how buildings will be used in order to achieve significant energy savings.
Link of the Week: Hobbyist Crafts Desktop Cray-1
As part of an ongoing “computational necromancy” project, electrical engineer Chris Fenton has knocked together a 1/10-scale, binary-compatible, cycle-accurate Cray-1. “The Cray-1 is one of those iconic machines that just makes you say ‘Now that’s a super computer!’” Fenton writes. “Sure, your iPhone is 10X faster, and it’s completely useless to own one, but admit it ... you really want one, don’t you?” The FPGA-powered scale model even includes the pleather couch-seat. “What’s the point of owning a Cray-1 if it doesn’t *look* like a Cray-1??” More from HPCwire.
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.