InTheLoop | 04.07.2008
The weekly newsletter for Berkeley Lab Computing Sciences
April 7, 2008
DOE SC Has Two Open Positions for Interdisciplinary Scientists
Patricia Dehmer, Deputy Director for Science Programs in the DOE Office of Science, has announced two interdisciplinary scientist vacancies. Applications are due by May 1, 2008. Here is the announcement:
I want to make you aware of two vacancies for scientists in the Office of the Deputy for Science Programs (SC-2), Office of Science, which are now advertised on USA Jobs. We are looking for two individuals with scientific backgrounds in any of the disciplines supported by the Office of Science. These individuals will serve a wide variety of functions, including program coordination, liaison, and integration within SC, across DOE, and across the federal government. The SC-2 office is young and growing, and we are looking for individuals who will help shape the office.
Because of the breadth of the programs in SC, we have advertised seven different positions with the following lead disciplines: Biologist, General Engineer, Physical Scientist, Physicist, Chemist, Mathematician, and Computer Scientist. The positions are advertised at the GS-13, GS-14, and GS-15 levels. Two individuals will be chosen from among those who apply for these positions. We are casting a wide net. Links to the positions are given below.
Biologist, GS-401-13/14/15, DE-SC-HQ-073 (kd):
General Engineer, GS-801-13/14/15, DE-SC-HQ-074 (kd):
Physical Scientist, GS-1301-13/14/15, DE-SC-HQ-075 (kd):
Physicist, GS-1310-13/14/15, DE-SC-HQ-076 (kd):
Chemist, GS-1320-13/14/15, DE-SC-HQ-077 (kd):
Mathematician, GS-1520-13/14/15, DE-SC-HQ-078 (kd):
Computer Scientist, GS-1550-13/14/15, DE-SC-HQ-079 (kd):
Future Technologies Group and ESnet Have Job Openings
The Future Technologies Group in the Computational Research Division has five job openings:
1. Research Scientist: Participate in research projects to develop methods and tools for performance evaluation and performance modeling of HPC computer systems and applications. These projects will involve a variety of activities, including: development of synthetic, serial and parallel benchmarks and associated performance models; benchmarking systems from single cores to large scale HPC systems with kernels and full applications; performance characterization, evaluation, modeling, and tuning of applications relevant to NERSC; developing auto-tuning software components of interest to FTG. Details: http://jobs.lbl.gov/LBNLCareers/details.asp?jid=21611&p=1.
2–3. Postdoctoral Researcher (two positions): Participate in research projects to develop methods and tools for performance evaluation and performance modeling of HPC computer systems and applications (as described in the Research Scientist position above). Details: http://jobs.lbl.gov/LBNLCareers/details.asp?jid=21612&p=1.
4-5. Computer Systems Engineer II and III (two positions): Participate in research projects to develop methods and tools for performance evaluation and performance modeling of HPC computer systems and applications (as described in the Research Scientist position above). The Computer Systems Engineer III will also lead complex software development efforts, including identifying requirements, proposing functionality, and overseeing actual development work. Details: http://jobs.lbl.gov/LBNLCareers/details.asp?jid=21613&p=1.
ESnet has an opening for a Unix Systems Administrator. As a member of ESnet’s Infrastructure Group, this position will be responsible for managing, configuring, and testing hardware and OS installations of all ESnet UNIX office and server machines with a special emphasis on the UNIX systems that are deployed at remote sites in support of ESnet’s production wide area network. The UNIX Systems Administrator will work closely with customers to define, develop, and implement enhancements. In addition, this role will implement and use tools and procedures to improve the security of systems, networks, and operational procedures. Details: http://jobs.lbl.gov/LBNLCareers/details.asp?jid=21635&p=1.
For information on the Lab’s Employee Referral Incentive Program (ERIP), go to http://www.lbl.gov/Workplace/HumanResources/ERIP/index_erip.html.
Seminar Today Will Explore Expanding Future Technologies Universe
Allan Snavely, Director of the Performance Modeling and Characterization Laboratory at the San Diego Supercomputer Center, will give a talk on “Red Shift and the Expanding Future Technologies Universe” today (Monday, April 7) at 1:00 p.m. in the 50F-1647 conference room. Here is the abstract:
This talk is about the future of computer technology as it encounters three basic laws:
1) “Einsteins Law,” which says that the speed of light is effectively constant.
2) Amdahl’s Law, which says that if you speed up only part of a multipart task the parts you do not speed up become critical-path.
3) Moore’s Law, often misquoted to suggest computers double in speed every 18 months but really just noting that historically transistor densities doubled at this frequency.
These three basic principles of physics, math, and technology underlie a phenomenon known as “Red Shift” whereby, analogous to our expanding universe, CPUs recede from their data with every turn of Moore’s Law. The further implications are profound. Without clever amelioration, CPUs just sit idle, dissipating more energy and waiting longer for their data with every turn of Moore’s Law; data movement, not computation per se, becomes the dominant bottleneck to performance (the Von Neumann bottleneck gets worse all the time); the energy cost of data motion in computation becomes a significant component of new electrical power brought online nationally; an exascale system, one a thousand times more powerful than today’s supercomputers, seems by straightforward extrapolation to need 1 GW of power. An empirical framework for measuring and quantifying the impact of Red Shift is presented; then it is shown how precision and scientific method embodied in performance modeling allows one to reason about and improve performance and efficiency of calculations at scale across subdisciplines of high performance computing including architecture design, data center deployment, and enabling heroic-sized scientific simulation.
The talk concludes with some technology forecasts: for example the rush to multicore is called out as being overly exuberant in the face of simple facts such as the three laws above.
Horst’s Lecture on Greening of HPC Will Be Webcast from TACC
As part of the Distinguished Lecture Series in Petascale Simulation sponsored by the University of Texas at Austin’s Institute for Computational Engineering and Sciences and the Texas Advanced Computing Center (TACC), Horst Simon will talk on “The Greening of HPC — Will Power Consumption Become the Limiting Factor for Future Growth in HPC?” from 3:30–5:00 p.m. Central Daylight Time on Thursday, April 10. Here is the abstract:
In a recent survey by IDC, facilities managers by an overwhelming majority named power and cooling to be the most pressing issues to them. A study of Exaflops computing concluded that by projecting today’s technology, an Exaflops computer might require 120 megawatts of power. A different study commissioned by the EPA estimates that server power consumption doubled in the period from 2000 to 2005 worldwide, and that the total amount of electricity consumed by servers worldwide now costs about $7.2B. This is the same order of magnitude as the investment in today’s HPC technology ($9.2B). Thus, we have reached a critical threshold that gives us cause to consider the question of power consumption as a potentially limiting factor to the future growth in HPC.
What are the power limitations of current technology, and how can we change the equation to assure the future rapid growth of HPC performance without contributing even more to carbon emissions and global warming? Dr. Simon will discuss several research projects that have started in Berkeley to address the issue of reducing power consumption in HPC, both at the systems and at the building level.
The lecture will be webcast live from http://livewebcast.theacesbuilding.com/. To view the webcast, get the browser plug-in from http://theacesbuilding.com/building/webcast/UTivo/Natural_Sciences/envivio_Intro_kbh.swf. Archived content from this lecture series is available at http://petascale.theacesbuilding.com.
Registration Is Open for SciDAC 2008 Tutorials
A day of multi-tracked tutorials will be held July 18 in Redmond, Washington following the main SciDAC 2008 meeting, July 13–17. The tutorials will provide instruction and hands-on practice with SciDAC-supported technologies. This year’s tutorials are sponsored in part by Microsoft Research, who will be providing transportation from the main conference site, the Fairmont Olympic, to the Redmond campus.
Berkeley Lab researchers will be leading four of the twelve tutorials. From CRD, John Wu will present “Accelerating Data Selection for On-Line Analysis with FastBit Indexing” for the SciDAC Scientific Data Management Center (SDM); Brian Tierney will present “Distributed Data Intensive Computing Tools” along with three colleagues from the Center for Enabling Distributed Petascale Science (CEDPS); and Tony Drummond will lead the “ACTS Toolkit Tutorial” for ACTS and TAU. HPC consultants from NERSC and Oak Ridge National Laboratory will present “An Introduction to the Cray XT4 for Application Scientists.”
NERSC’s David Skinner organized the tutorials in his role as head of the SciDAC Outreach Center.
Full abstracts and registration are available online at https://outreach.scidac.gov/scidac08/tutorials/.
Grace Hopper Celebration Call for Participation Closes April 15
The Grace Hopper Celebration of Women in Computing will be held at the Keystone Resort in Colorado from October 1–4, 2008. The series of conferences is designed to bring the research and career interests of women in computing to the forefront. Presenters are leaders in their respective fields, representing industrial, academic, and government communities. Leading researchers present their current work, while special sessions focus on the role of women in today’s technology fields, including computer science, information technology, research and engineering.
The submission deadline for all programs except panels and workshops has been extended to April 15, 2008. For more information on the conference, go to http://gracehopper.org/2008/. The call for participation is at http://gracehopper.org/2008/participate/call-for-participation/.
Anita Borg and Denice Denton Awards Are Open for Nomination
The Anita Borg Institute for Women and Technology has announced that the Anita Borg and Denice Denton Awards are open for nomination. Nominations must be submitted by April 30, 2008.
The Denice Denton Emerging Leader Award (http://anitaborg.org/initiatives/awards/denice-denton-award/) recognizes a junior (under 40), non-tenured faculty individual in an academic or research institution who is pursuing high quality research in engineering or physical sciences, while contributing significantly to promote diversity in his/her environment. The award winner will receive a $5,000 prize underwritten by Microsoft.
The Anita Borg Social Impact Award (http://anitaborg.org/initiatives/awards/anita-borg-awards/2007/anita-borg-social-impact-award/) will honor an individual or team who has caused technology to have a positive impact on the lives of women and society, or who has caused women to have a significant impact on the design and use of technology. The award recipient will receive a $10,000 prize underwritten by Microsoft.
The Anita Borg Technical Leadership Award (http://anitaborg.org/initiatives/awards/anita-borg-awards/2007/anita-borg-technical-leadership-award/) will recognize and celebrate an outstanding woman technical leader. Nominees are women who have inspired the women’s technology community through outstanding technological and social contributions and, through their leadership, have increased the impact of women on technology. The award recipient will receive a $10,000 prize.
Go to https://ssl.linklings.net/organizations/abi/ to nominate someone for these awards. Awards will be distributed at the Grace Hopper Celebration of Women in Computing in October.
Research Opportunities Offered for Undergraduate Women and Minorities
The Computing Research Association’s Committee on the Status of Women in Computing Research (CRA-W), in conjunction with the Coalition to Diversify Computing (CDC), has announced two programs that involve undergraduate student research. The goal of these initiatives is to increase the numbers of women and minorities who continue on to graduate school in computer science and engineering.
The Collaborative Research Experience for Undergraduates (CREU) program and the Multidisciplinary Research Opportunities for Women (MRO-W) program are designed to provide positive research experiences for teams of undergraduates working at their home institutions. Students will work with one or two sponsoring faculty members on a project for which monetary support is typically not available. Students will each receive a stipend of $3,000 for their work during the academic year and $4000 for their MRO-W summer work. CREU may also receive $500 and MRO-W may receive $1500 to be used for associated special equipment, travel, or supporting materials.
See the websites for specific information about eligibility, evaluation criteria, and applying:
- CREU website: http://www.cra.org/Activities/craw/creu/
- MRO-W website: http://www.cra.org/Activities/craw/mrow/
Application deadline is May 9, 2008. This initiative is offered in cooperation with the National Science Foundation.
Link of the Week: Are Our Brains Wired for Math?
According to neuroscientist Stanislas Dehaene, humans have an inbuilt “number sense” capable of some basic calculations and estimates. The problems start when we learn mathematics and have to perform procedures that are anything but instinctive. Exact calculations are difficult because they involve three different parts of the brain. Read the New Yorker article at http://www.newyorker.com/reporting/2008/03/03/080303fa_fact_holt.
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.