UC Regents Approve New Management Contracts for Three DOE Labs
The University of California Board of Regents last week approved five-year contract extensions for UC's management of three DOE national laboratories. The contracts--one each for Berkeley Lab, Lawrence Livermore and Los Alamos National Lab--are the result of more than a year of negotiations between the university and the DOE. The contracts, which run concurrently through September 2002, were signed by UC and DOE officials after the Regents' meeting in San Francisco.
The university has been asked by the federal government to manage thethree laboratories since their inception. The partnership began in 1943, with the establishment of the Berkeley laboratory as a federal facility--it is the first laboratory in the DOE complex--and the founding of the Los Alamos laboratory that same year; the Livermore laboratory was founded in 1952. The laboratories employ a combined UC workforce of more than 17,000 people and operate on federally funded budgets totaling about $2.4 billion.
Who's Who in Computing Sciences: Eric Essman
When Eric Essman packs up his shelves for his pending move to Bldg. 50F, the moving boxes will fill with titles testifying to his varied interests. There's "Faust," "The Oxford Dictionary of Opera," "Godard par Godard," Einstein's "Relativity," "Galileo Heretic," "Classical Myth in Literature, Art and Music," and "1001 Pitfalls in French," among many others. Eric arrived at the Lab 14 years ago, applying for a job here after watching a series of Richard Feynman's physics lectures on TV. With a bachelor's in English and master's in English lit, both from S.F. State, Eric admits the science and math side of his brain was semi-atrophied at the time. To remedy that condition, he took some math classes and began reading science magazines.
He started out at the Lab with the Applied Sciences Division and has also worked in Accelerator and Fusion Research, the Center for X-ray Optics, and the National Center for Electron Microscopy. With the arrival of NERSC, Eric saw a chance to be part of a very dynamic program and signed on. He provides admin support for the UNIX group, and the User Services, Scientific Computing and the Visualization groups in NERSC.
Citing philosopher Isaiah Berlin, who divides the world into foxes (generalists) and hedgehogs (specialists), Eric says he comes down on the fox side. This trait, he says, led him to study 17th century English literature. "That period was the last time in Western history when one person could know everything relevant about all disciplines, from science to the humanities. The best minds of that period could synthesize history, the Bible, science, politics and law in a way that's nearly impossible today."
A native Southern Californian--a beach-baby if not a Beach Boy--born in Santa Monica, Eric says he also developed a strong attachment to San Francisco by visiting his grandmother in Pacific Heights every summer. When it came time to go to college, he chose S.F. State, pursuing degrees first in radio and TV broadcasting, then history, before settling on English. He met his wife at college and today they are both avid film followers. She is chair of the film department at the University of the Pacific in Stockton and Eric coordinates an annual film festival in San Francisco. His all-time top film recommendation? Hitchcock's "Vertigo."
Workshop on Building Computational Sciences at Berkeley Lab
Last week, a two-day workshop was held by NERSC to assess efforts to create an intellectual foundation for computational science-of-scale as a permanent research activity at Berkeley Lab, as well as make the Lab a preeminent location for using supercomputing machines and techniques to solve problems at the forefront of scientific research. According to NERSC Division Leader Horst Simon, who organized the meeting, a key goal of the meeting in Oakland was to establish two-way communication between Computing Sciences/NERSC researchers and developers and the Lab community at the principal investigator/senior scientist level.
Lab researchers heard updates on the computational and computer science capabilities Computing Sciences has to offer, while CS staff had an opportunity to find out about new potential applications of the tools and technologies currently under development, as well as about unsolved computational science challenges. After a series of presentations, participants broke into working groups to discuss such issues as:
- What are the significant problems in Computational Science which we can address at LBNL?
- What are our strategic advantages, such as the ALS, Human Genome Project etc.?
- Where can we gain maximal advantage by combining computer science research, NERSC systems, computer science expertise on campus with LBNL scientific programs?
- How can we go about building computational science programs, and assure long-term funding?
Working groups were formed in the areas of High Energy Nuclear Physics (led by Jim Siegrist, JLSiegrist@lbl.gov), Material Sciences (Bill McCurdy, CWMcCurdy@lbl.gov), Life Sciences (Teresa Head-Gordon, TLHead-Gordon@lbl.gov), and Earth Sciences and Energy and Environment (Karsten Pruess, K_Pruess@lbl.gov, and Nancy Brown, NJBrown@lbl.gov). Among the identified hurdles are better data management, bridging experimental and computational science, and giving researchers a clearer understanding of what NERSC and Computing Sciences can provide.
The material sciences, life sciences, and earth sciences groups agreed to develop white papers on how to build computational sciences efforts in these Lab areas. The physics/nuclear sciences/AFRD group decided to focus on data intensive computing and is looking to draft a document to enhance NERSC's capability in this area. Plans are to present the papers and results of further study at a follow-up workshop in January. For further questions about the follow-up activities, or if you are interested participating in the workgroup, please contact the discussion leaders.
The featured speaker was Paul Saffo of the non-profit Institute for the Future in Menlo Park. Saffo describes himself as a historian of technology who studies technologies that don't exist yet. Focusing on a theme of how inexpensive technologies affect society, Saffo pointed to semiconductors creating the PC revolution in the 1980s, lasers shaping the '90s and predicted that sensors would be the technological magic wand of the coming decade. Combined with electronic activating devices, electronic "eyes and ears" will play important roles in our future lives. A current, but not too-reliable, example are auto airbags, Saffo says. He predicts changes in everything from transportation to medicine to communications to warfare. To read more of his ideas, visit the web site at: http://www.iftf.org/sensors/sensors.html
Who You Gonna Call? 486-HELP
The Lab's new computer support "help" desk is shaping up with a new phone number, new quarters and new employees. Part of the Computing Infrastructure Support Department, the help desk is being developed as a single point of contact for all desktop computer support issues, from hardware to software to networking problems. The desk can be reached by dialing H-E-L-P (4357). The staff, currently made up of Marty Gelbaum, Marty Morimoto and Sheri Palmer, are settling into new cubicles in the revamped Bldg. 50F.
SC97: High Performance Networking and Computing Set for Nov. 15-21
SC97, sponsored by ACM and IEEE, brings together scientists, engineers, designers and managers from all areas of high-performance networking and computing and showcases the latest in systems, applications and services. Highlights include the presentation of technical papers, tutorials, education sessions, research exhibits, poster exhibits and exhibits from industry.
A new feature, state-of-the-field talks, will include invited speakers discussing the current state of work in their field, and presenting problems which may be solved with input from outside disciplines. State-of-the-Field speakers include David Farber, Alfred Fitler Moore Professor of Telecommunications at the University of Pennsylvania and member of the Presidential Advisory Committee on HPCC, IT and NGI; Pat Hanrahan, Canon USA professor of computer science and electrical engineering at Stanford University; John L. Hennessy, co-founder of MIPS Computers Systems and dean of engineering at Stanford University; and Ken Kennedy, co-chair of the Presidential Advisory Committee on HPCC, IT and NGI and director of the Center for Research on Parallel Computation.
For more info, see http://www.supercomp.org/sc97 or write email@example.com to be added to the mailing list to receive the SC97 Advance Program.
Questions, Comments About InTheLoop?
Send your questions, comments, suggestions or news items to JBashor@lbl.gov or call editor Jon Bashor at 5849.