![]() |
|
||||||
| Home
| Mission
|
about SciDAC
|
Contact Us |
||||||
Archive of SciDAC News and NotesSciDAC in the News (press releases and media mentions) December 2009November 2009DOE labs sweep major awards at SC09 conferenceORNL-Led Team Takes Gordon Bell Prize for World’s Fastest Science AppA team led by ORNL’s Markus Eisenbach was named winner of the 2009 ACM Gordon Bell Prize, which honors the world’s highest-performing scientific computing applications. Another team led by ORNL’s Edo Aprà was also among nine finalists for the prize. Results of the contest were announced in Portland, Oregon, during the SC09 international supercomputing conference. The prize is supported by high-performance computing pioneer Gordon Bell and is administered by the Association for Computing Machinery. IBM-LBNL Simulation of Cat-Size Cortex is a Gordon Bell Prize Special Category WinnerA team of researchers from the IBM Almaden Research Center and the Lawrence Berkeley National Laboratory won the prestigious Gordon Bell Prize in the special category for their development of innovative techniques that produce new levels of performance on a real application. This year's prize winners were announced Thursday, Nov. 19, 2009 at the awards session of the SC09 conference in Portland. The ACM Gordon Bell Prize annually recognizes the best performance of scientific applications on supercomputers. more Tennessee Supercomputing Titans Triumph at SC09 HPC Challenge AwardsTwo powerful Cray XT5 systems at the ORNL computing complex outmuscled competitors to win half of this year’s High-Performance Computing (HPC) Challenge awards. Results of the “Best Performance” awards, which measure excellence in handling computing workloads, were announced Nov. 17 at SC09. ORNL’s Jaguar took home the lion’s share of the honors, with three gold medals and one bronze. Kraken, an academic supercomputer located at ORNL and funded by the National Science Foundation (NSF) through a partnership with the University of Tennessee, showed that it too is a contender with two silver medals. Jaguar won first place for speed in solving a dense matrix of linear algebra equations by running the HPL software code at 1,533 teraflop/s (trillion floating point operations per second). Kraken, the world’s fastest academic computer, took second by running HPL at 736 teraflop/s. Jaguar also ranked first for sustainable memory bandwidth by running the STREAM code at 398 terabytes per second. STREAM measures how fast a node can fetch and store information. Jaguar’s third gold was for executing the Fast Fourier Transformation (FFT), a common algorithm used in many scientific applications, at 11 teraflop/s. Kraken took second with a speed of 8 teraflop/s. SDSC, UC San Diego, LBNL Team Wins SC09 'Storage Challenge' AwardA research team from the San Diego Supercomputer Center (SDSC) at UC San Diego and the University of California's Lawrence Berkeley National Laboratory (Berkeley Lab) has won the Storage Challenge competition at SC09, the leading international conference on high-performance computing, networking, storage and analysis held Nov. 14-20 in Portland, Oregon. The research team based its Storage Challenge submission for the annual conference on the architecture of SDSC's recently announced Dash high-performance computer system, a "super-sized" version of flash memory-based devices such as laptops, digital cameras and thumb drives that also employs vSMP Foundation software from ScaleMP, Inc. to provide virtual symmetric multiprocessing capabilities. Berkeley Lab's Peter Nugent and Janet Jacobsen provided the project a production database that they could use for the data challenge, and did so in a short time. more Extreme Scale Workshops
Presentations are posted from SciDAC 2009 held June 14-18, 2009. October 2009Extreme Scale Workshops
PETSc named to R&D 100
September 2009Speeding up Data Transfers: Bigger Pipes between NERSC and OLCF
Network systems engineers from the Department of Energy’s (DOE) Energy Sciences Network (ESnet), National Energy Research Scientific Computing Center (NERSC) and Oak Ridge Leadership Computing Facility (OLCF) are teaming up to optimize wide-area network (WAN) data transfers. Oak Ridge Leadership Computing Facility (OLCF), located in the National Center for Computational Science (NCCS) at Oak Ridge National Laboratory in Tennessee, and NERSC, located at Lawrence Berkeley National Laboratory in California, are home to some of the fastest supercomputers in the world. A number of research groups use resources at both centers. ESnet, DOE’s high-speed network, connects the two centers, as well as other national labs and universities around the country. With the installation and deployment of new dedicated data transfer nodes at NERSC and OLCF linked by ESnet, researchers are now able to move large data sets between each facility’s mass storage systems at a rate of 200 megabytes per second (MB/sec). At this rate, 74 terabytes of information in the U.S. Library of Congress’ digital collection could be transferred in approximately four days. more July 2009
June 2009
SciDAC’s VACET Team Demonstrates Tools for Analyzing Massive Datasets
As computational scientists are confronted with increasingly massive datasets from supercomputing simulations and experiments, one of the biggest challenges is having the right tools to gain scientific insight from the data. A team of DOE researchers recently ran a series of experiments to determine whether VisIt, a leading scientific visualization application, is up to the challenge. Running on some of the world’s most powerful supercomputers, VisIt achieved unprecedented levels of performance in these highly parallel environments, tackling data sets far larger than scientists are currently producing. The team ran VisIt using 8,000 to 32,000 processing cores to tackle datasets ranging from 500 billion to 2 trillion zones, or grid points. The project was a collaboration among leading visualization researchers from Lawrence Berkeley National Laboratory, Lawrence Livermore National Laboratory, and Oak Ridge National Laboratory. Specifically, the team verified that VisIt could take advantage of the growing number of cores powering the world’s most advanced supercomputers, using them to tackle unprecedentedly large problems. Scientists confronted with massive datasets rely on data analysis and visualization software such as VisIt to “get the science out of the data,” as one researcher said. VisIt, a parallel visualization and analysis tool that won an R&D 100 award in 2005, was developed at LLNL for the National Nuclear Security Administration. more March 2009
January 2009Extreme Scale Workshop Series
December 2008
November 2008October 2008
September 2008
August 2008
July 2008June 2008Argonne's Blue Gene/P Named World's Fastest for Open Science
While the Blue Gene/P has a peak-performance of 557 Teraflops, Intrepid achieved a speed of 450.3 Teraflops on the Linpack application used to measure speed for the Top500 rankings. "Intrepid's speed and power reflect the DOE Office of Science's determined effort to provide the research and development community with powerful tools that enable them to make innovative and high-impact science and engineering breakthroughs," said Rick Stevens, associate laboratory director for computing, environmental and life sciences at Argonne. "Scientists and society are already benefitting from ALCF resources," said Peter Beckman, ALCF acting director. "For example, ALCF's Blue Gene resources have allowed researchers to make major strides in evaluating the molecular and environmental features that may lead to the clinical diagnosis of Parkinson's disease and Lewy body dementia, as well as to simulate materials and designs that are important to the safe and reliable use of nuclear energy plants." Eighty-percent of Intrepid's computing time has been set aside for open science research through the DOE Office of Science's (SC) highly select Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program. There are currently 20 INCITE projects at the ALCF that will use 111 million hours of computing time this year. SC's Office of Advanced Scientific Computing Research provides high-level computer power focused on large-scale installation used by scientists and engineers in many disciplines. For more details, see the "Argonne Press Release" May 2008The Open Science Grid(OSG) Consortium: Collaborative Science over the Grid
The Open Science Grid (OSG) is one of several SciDAC-2 projects that are at the forefront of the continuing United States investments in Grids. OSG itself is leading in the provision of a community based, high-throughput nationally distributed infrastructure for a broad sweep of scientific research. Today OSG provides access to computational resources at more than sixty sites across the US – including NERSC, Brookhaven National Laboratory and Fermilab, more than six campus-wide university infrastructures, and many large and small university sites across the country. OSG relies on today's state of the art networks for science, and advanced experimental research networks to support high throughput data distribution and computing for the experiments at the Large Hadron Collider at CERN, existing DOE Office of Science physics experiments, bioinformatics, chemistry, climate research, computer science and other domains. The infrastructure today is accepting more than thirty percent of the data and providing more than twenty-five percent of the analysis throughput for the LHC. Through the Virtual Data Toolkit, OSG supplies an integrated suite of software technologies used on the Grid based on the core Condor and Globus technologies created by the “fathers” of the Grid at Universities in the US, and augmented by tools and utilities from many places including Europe. OSG is supported by the Department of Energy’s Office of Science SciDAC-2 program from the High Energy Physics, Nuclear Physics and Advanced Software and Computing Research programs; and the National Science Foundation Math and Physical Sciences, Office of Cyber-Infrastructure and Office of International Science and Engineering Directorates.
Contact Ruth Pordes (ruth@fnal.gov) or Miron Livny(miron@cs.wisc.edu) for more details.
URL: www.opensciencegrid.org April 2008
"Exascale" Town Meeting Report is online
The objective of this ten-year vision is to focus the computational science experiences gained over the past ten years on the opportunities introduced with exascale computing to revolutionize our approaches to energy, environmental sustainability and security global challenges.
March 2008
SciDAC Review Spring issue now onlineThe Spring 2008 issue includes an interview with LBNL Director Dr. Steve Chu about collaborations for the future. Project articles include Fusion, Subsurface flow, and Climate. Other stories cover ANL, determining optimal simulation size, and better meshes for accelerators. Accomplishments, meetings, and workshops are highlighted. more January 2008SciDAC well represented in INCITE Awards
Of the 55 projects awarded computer time through the INCITE program for 2008, 15 are current SciDAC projects and 11 are to researchers from former SciDAC projects. Although SciDAC projects represent only 27% of the projects awarded, they account for 38% of the total processor hours awarded -- nearly 102 million of the over 267 million processor hours. Top awards to current SciDAC projects include 26.7 million processor hours for Lattice Quantum ChromoDynamics (LQCD) (R. Sugar, UCSB), 18 million for climate modeling (W. Washington, NCAR), and 17.5 million for nuclear structure research (D. Dean, ORNL). Top awards to researchers from past projects include 18 million processor hours for astrophysics (A. Mezzacappa, ORNL) and 14 million for applied math (P. Fischer, ANL). The Innovative and Novel Computational Impact on Theory and Experiment (INCITE) program was conceived specifically to seek out computationally intensive, large-scale research projects with the potential to significantly advance key areas in science and engineering. The program encourages proposals from universities, other research institutions and industry. For 2008, computational resources for the INCITE program will be provided by Argonne, Lawrence Berkeley, Oak Ridge and Pacific Northwest national laboratories. 2008 INCITE fact sheets (pdf) SciDAC Outreach Center Is Soliciting Tutorials for SciDAC 2008A day of tutorials will be held in conjunction with this year's SciDAC conference (SciDAC 2008). The main meeting will be held July 13-17 at the Fairmont Olympic Hotel in Seattle, and the tutorial day will be held in Redmond on Friday the 18th. Transportation will be provided from the conference hotel to the tutorial site. This year's tutorials are sponsored by Microsoft Research and the SciDAC Outreach Center. The audience for the tutorials is expected to include students from the surrounding universities, DOE Computational Science Graduate Fellowship students, and interested parties from industry. Last year there were just under a hundred students participating. The SciDAC Outreach Center is soliciting ideas for tutorials from the SciDAC PIs. Anyone who would like to give a tutorial is asked to send the proposed topic to David Skinner (DESkinner@lbl.gov), along with any AV or computing requirements your tutorial might require. Projectors, wireless network, etc. will be provided. As they did last year, the SciDAC Outreach Center can provide temporary training accounts and batch queues for tutorials that would benefit from hands-on access to XT4, POWER5, and other HPC architectures. The 2007 tutorial agenda is still available (for your reference). December 2007
November 2007
CTWatch Quarterly Publishes All-SciDAC IssueThe November 2007 issue of CTWatch Quarterly is titled "Software Enabling Technologies for Petascale Science". The issue is comprised of papers from SciDAC Institutes and Centers, with an introduction by Fred Johnson of DOE. Articles in this issue are: DOE and SciDAC at SC07When the world’s leading experts in high performance computing, networking, storage and analysis convene in Reno, Nevada November 10-16, much of the scientific and technical program will reflect the expertise of researchers sponsored by the Department of Energy (DOE) at national laboratories and universities across the country. Most notably are two of the invited plenary talks on Wednesday in room C-4 of the Reno-Sparks Convention Center. At 8:30 a.m. DOE Under Secretary for Science Raymond L. Orbach will speak on “The American Competitiveness Initiative: Role of High End Computation”. Dr. Orbach will be followed at 9:15 a.m. by 2006 Nobel Laureate George Smoot, astrophysicist at DOE’s Lawrence Berkeley National Laboratory, who will talk on “Cosmology's Present and Future Computational Challenges”. Throughout the course of the four-day SC07 technical program, DOE-supported researchers will have a role in: 18 of the 54 peer-reviewed technical papers, 5 of the 7 panel discussions, 10 of the 25 tutorial sessions, 6 of the 10 workshops, 2 of the 15 invited “Masterworks” talks, 13 of the 45 research posters. Also, the prestigious Sidney Fernbach Computer Science and Engineering Award will be presented to Columbia University Prof. David Keyes, the leader of a DOE project to develop software to more efficiently solve problems involving partial differential equations. The project is part of DOE’s Scientific Discovery through Advanced Computing (SciDAC program). October 2007More projects added to SciDAC-2Improving Ethanol Production Efficiency: Understanding Cellulase Enzymes
September 2007More projects added to SciDAC-2Pointing the Way for Accelerator Science: ComPASS Project Aims for Petascale
July 2007
As a complement to the SciDAC 2007 Conference June 24-28, a hands-on SciDAC Tutorials Workshop will be held June 29 at MIT. The SciDAC Tutorials Workshop is open to all physical scientists, mathematicians, computer scientists, and computational scientists, and we anticipate participation both by attendees of SciDAC 2007 and local students, postdocs, and researchers. There is no workshop fee, but registration in advance is required, and in the event of oversubscription, space in the tutorials will be allocated according to the order of registration. Participants should bring their own laptops for wireless access the tutorial software. The SciDAC Tutorials Workshop is sponsored by:
June 2007SciDAC Review Summer issue now online
SciDAC Review is a quarterly publication dedicated to sharing information, news, and achievements relevant to the SciDAC program and computational science in general. 2007 SciDAC Conference Draws Over 300 Researchers
Images (clockwise from top): Plenary session, Vendor response to TownHall, Program Managers (3), Poster session
More than 300 members of the computational science community met in Boston June 24-28 at the 2007 SciDAC Conference. In welcoming participants, SciDAC Director Michael Strayer noted that “Perhaps no other area of research of the Department of Energy is so critical as computational science for understanding the fundamental components of nature and the behavior of complex systems.” The conference program featured 36 plenary talks covering research areas ranging from nanotechnology to massive exploding stars, from climate change to developing fusion reactors as future energy sources. Other talks provided insight into the challenges of effectively utilizing the complex architectures of supercomputers for scientific discovery. Conference participants, representing 55 universities, 20 national laboratories, 14 private companies, five government agencies and four countries also attended two evening poster sessions, during which 76 research posters were presented, often generating in-depth discussions of further research possibilities. Such interactions are a hallmark of the SciDAC program, which brings together multi-disciplinary teams from different institutions to tackle some of the most pressing scientific problems. ![]() March 2007
Annual Conference in Boston
New SciDAC grants have been posted in four Fusion Energy areas: [FAPN07-19 | LAB07_19] [FAPN07-20 | LAB07_20] [FAPN07-21 | LAB07_21] [FAPN07-22 | LAB07_22] February 2007FY07 Appropriation PassedThe SciDAC appropriations for FY07 were passed by congress and signed by the President in mid February. Anyone whose funding was caught up in the Continuing Resolution should see movement very soon.
Nearly 80 SciDAC-supported researchers, project managers and support staff gathered in Atlanta in February to reinforce existing partnerships and discuss potential new areas of collaboration. The day-and-a-half meeting was organized by the SciDAC Outreach Center as part of its mission to support SciDAC outreach, training and research objectives. A primary focus of the outreach center is to foster communication within the high performance computing community. In addition to presentations about collaborations between scientists, applied mathematicians and computer scientists under the first set of SciDAC projects, the Atlanta meeting also featured two poster sessions. The posters provided talking points to get representatives of the SciDAC-2 scientific applications, centers and institutes talking. As a result, groups of various sizes congregated around the posters, talking intently about possible collaborations and support. “Without the posters, it would have been very hard, if not impossible, for us to communicate with one another about some of the key concepts, challenges, and capabilities,” said Kwan-Liu Ma, leader of the SciDAC Institute for Ultrascale Visualization. “The posters also enabled us to quickly identify potential collaborations and limiting the meeting size made it possible for me to get to know and also talk to almost all the PIs. It was also helpful to the answers of many questions asked by others but I did not think of.” One interactive poster featured a matrix in which PIs were asked to identify known and potential collaborations. The data collected from this poster will be used to create a Web-based display of SciDAC partnerships, which are an integral component of the program’s strategy for achieving computational science breakthroughs. January 2007SciDAC PI Organizational Meeting, February 5-6 in AtlantaFirst Annual Cray Technical Workshop - USA, February 26-28 in NashvilleOctober 2006New SciDAC SolicitationsThe DOE Office of Science has issued two new solicitations to universities for SciDAC-2. The research areas are the Climate Change Prediction Program and Accelerator Science and Simulation. Accelerator Science and Simulation is also the topic of a related program announcement to DOE National Laboratories. September 2006Science at the PetascaleOn September 7, 2006 DOE announced $60 Million in new SciDAC projects (DOE press release | list of projects). Scientific Discovery through Advanced Computing (SciDAC) computational science projects are aimed at accelerating scientific research in designing new materials, developing future energy sources, studying global climate change, improving environmental cleanup methods and understanding physics from the tiniest particles to massive exploding stars known as supernovae. As most of the projects funded under the initial SciDAC program have officially ended (alumni projects on this site) and the resulting success stories (pdf) underscore the value of multi-disciplinary collaborations. SciDAC-1 might appear to be a very hard act to follow. But with the announcement of SciDAC-2 projects, it’s clear that the Department of Energy (DOE) and the nation’s scientific computing community are up to the task – taking scientific advance to the petascale. [MORE]
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Home | ASCR | Contact Us | DOE disclaimer |
|
|