InTheLoop | 10.31.2005
The Weekly Electronic Newsletter for Berkeley Lab Computing Sciences Employees
October 31, 2005
CRD's Julian Borrill Joins U.S. Planck Internal Advisory Board
Julian Borrill, an astrophysicist in CRD's Scientific Computing Group, was recently invited and agreed to join the U.S. Planck Internal Advisory Board. The board is made up of half a dozen or so senior members of the U.S .Planck collaboration, including experimentalists, data analysts and theorists, and spanning the range of participating US institutions. Its role is to advise the two U.S .Planck PIs, Charles Lawrence at JPL and Andrew Lange at Caltech, on all aspects of the U.S. Planck program.
In the year 2007 the European Space Agency, with substantial NASA participation, will launch the Planck satellite on a mission to map the cosmic microwave background (CMB), the remnant radiation believed to be an echo of the Big Bang that started the universe. Planck is designed to map CMB temperature and polarization fluctuations with unprecedented resolution and sensitivity, but the enormous volume of data this will generate poses a major computational challenge.
A group led by Julian has already reached a major milestone in preparing for the onslaught of data. Late last year, the group processed an entire year's worth of simulated Planck data at the single most CMB-sensitive frequency, producing a high-resolution CMB map in just two hours. Using NERSC's Seaborg supercomputer, the researchers ran their data on 6,000 of Seaborg's 6,080 processors at once, mapping 75 billion observations to 150 million pixels. The run was the first ever to use virtually all of Seaborg's computing processors on a single code.
IT Division Rolls Out New Audio Conference Service at the Lab
In an effort to help reduce costs, Telephone Services is providing a new audio conference service to the Lab at a rate that is up to 70 percent lower per minute, effective Tuesday, November 1. In addition to the cost savings, users will now have the flexibility to schedule their own teleconferences any time of day (24 hours a day, seven days a week) without having to submit a request. Users are assigned a 20-party audio conference account with both toll free and toll numbers. International callers must use the toll number. Larger audio conference accounts can be set up, as needed.
To find out more about Lab conference call services, visit http://tscweb.lbl.gov/conference/conference_calls.html or call Telephone Services at ext. 7997.
NERSC's Systems Group Gets Up to Speed with New File System
As part of a run-up to the SC05 conference to be held next month in Seattle, Will Baird of NERSC's Computational Systems Group put the center's Global Parallel File System (GPFS) through its paces, achieving data transfer rates up to four times faster than previously. Will has entered the SC05 Tri-Challenge, a combination of three separate challenges HPC Analytics, StorCloud and the Bandwidth Challenge. The team that posts the highest combined score will be crowned the Tri-Challenge winner.
We had a very successful test of the GPFS 2.3 wide area functionality, Will wrote last week. We were able to do some tuning that allowed us to move the performance of transfers from between 1.5 MB/s to 1.6 MB/s to around 6 MB/s.
GPFS is the software being used to deploy NERSC's Facility-Wide File System (FWFS). The benefits of FWFS to scientific productivity are manifold. By providing a single unified namespace, FWFS will make it easier for users to manage their data across multiple systems. Users will no longer need to keep track of multiple copies of programs and data; they will no longer need to copy data between NERSC systems for pre- and post-processing. Storage utilization will become more efficient through decreased fragmentation. Computational resource utilization will become more efficient as users can more easily run jobs on an appropriate resource. FWFS will also provide improved methods of backing up user data.
These attributes also come into play in the Tri-Challenge, which involves moving huge amounts of data across the network, storing and retrieving vast quantities of data, and using leading-edge analytics techniques to solve complex, real-world problems.
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.