A-Z Index | Phone Book | Careers

InTheLoop | 12.13.2010

December 13, 2010

Berkeley Lab Team Wins Best Paper at CloudCom

Cloud computing has proven to be a cost-efficient model for many commercial web applications, but will it work for scientific computing? Not if the application heavily relies on interconnect communications, writes a team from Berkeley Lab. Their paper, titled “Performance Analysis of High Performance Computing Applications on the Amazon Web Services Cloud,” was honored with the Best Paper Award at the IEEE’s CloudCom 2010, held Nov. 30–Dec. 1 in Bloomington, Ind. Read more.

ESnet Will Add Flexible Reservation Algorithm to OSCARS

As scientific research becomes more collaborative and more data-intensive, larger teams of scientists are generating, sharing and analyzing increasingly large datasets. Many of these applications need networking support that provides predictable performance, which in turn requires effective algorithms for bandwidth reservations. A team of Berkeley Lab researchers have recently developed an algorithm that will allow users to inquire about bandwidth availability, and receive alternative suggestions when reservation requests fail. This flexible reservation algorithm for advance network provisioning will be integrated into the next version of ESnet OSCARS, version 0.6, due out in early 2011. Read more.

IBM Case Study Analyzes Carver and Magellan’s Energy Efficiency

A recently published IBM case study, “NERSC creates an ultra-efficient supercomputer,” describes how the Carver and Magellan clusters were designed to maximize utilization of space, power, and cooling capacity, resulting in a lower-cost, higher-performance option than the clusters they replaced. Brent Draney, NERSC’s group lead for Network, Security, and Servers, is quoted throughout the case study, describing the innovative design and its benefits for NERSC. Read more.

Network Matters: Why This Spiking Network Traffic?

Last month was the first in which the ESnet network crossed a major threshold — over 10 petabytes of traffic! Traffic volume was 40% higher than the prior month and 10 times higher than just a little over four years ago. But what’s behind this dramatic increase in network utilization? Read more.

ICCS Workshop in January on Many-Core Scientific Computing

Registration is now open for a workshop on Manycore and Accelerator-Based High-Performance Scientific Computing to be held January 24–28, 2011, in Berkeley, Calif. Workshop organizers have also issued a call for research papers and posters. The deadline for submissions and registration is Tuesday, January 4.

The workshop, now in its second year, is organized by the International Center for Computational Sciences (ICCS) located at Lawrence Berkeley National Laboratory and the University of California, Berkeley. ICCS is an international collaboration to research and deliver state-of-the-art high-performance computing (HPC) hardware and software solutions to broader scientific communities. Read more.

David Skinner Comments on New Globus Online Service

A new tool that will provide secure file–transfer service to manage large–scale data was introduced Nov. 18 at SC10 in New Orleans. Globus Online will use a cloud–based system rather than complex, custom information technology infrastructure to provide high performance file transfers. NERSC Outreach, Software, and Programming Group lead David Skinner comments on the service in a PhysOrg.com news article.

HPCS 2011 Issues Call for Papers and Participation

The 2011 High Performance Computing and Simulation Conference (HPCS 2011) will be held July 4–8, 2011 in Istanbul, Turkey. You are cordially invited to participate in this conference through paper submission, a workshop or a special session organization, a tutorial, an invited speech, a demo, a poster, an exhibit, a panel discussion, a doctoral dissertation, whichever sounds more appropriate and convenient to you. The paper submission deadline is January 11, 2011. For more information, see the website.

CITRIS Student Competition offers $45K in Prizes

The Center for Information Technology Research in the Interest of Society (CITRIS) has announced the sixth annual CITRIS Big Ideas white paper competition. This year CITRIS will give away $45K in cash prizes for the best ideas that demonstrate the ability of information technology to address a major societal challenge. The cash can be used as a scholarship or to support the proposed project.

The IT for Society contest is open to students from all four CITRIS campuses: UC Berkeley, UC Merced, UC Santa Cruz, and UC Davis. The deadline this year is Friday, March 7, 2011. Go here for contest information, and here to see last year’s winners.

This Week’s Computing Sciences Seminars

Visualizing Patterns in Art, Games, Comics, Photography, Cinema, Animation, and Print Media
Tuesday, December 14, 12:00–1:00 pm, OSF 943-238
Lev Manovich and Jeremy Douglass, Visual Arts Department, University of California, San Diego

The explosive growth of cultural content on the web, including social media and the digitization work by museums, libraries, and companies, makes possible a fundamentally new paradigm for the study of culture and media. We can use computer-based techniques for data analysis and interactive visualization already employed in sciences to analyze patterns and trends in massive cultural data sets. We call this paradigm Cultural Analytics.

We will show examples of our visualizations of patterns in art, film, animation, video games, magazines, literature and comics created in our lab (softwarestudies.com) situated at University of California, San Diego (UCSD) and California Institute of Telecommunication and Information (Calit2). We will also discuss in detail our work on the analysis and visualization of our 1 million Manga pages which used the supercomputers at NERSC (funded by NEH Humanities High Performance Computing program) and 215 megapixel HIPerSpace display developed at Calit2.

Implicit Shape Representations and Analysis in CT Liver Segmentation
Thursday, December 16, 11:00 am–12:00 pm, 50F-1647
Grace Vesom, Department of Engineering Science University of Oxford, UK

Liver segmentation is a longstanding research endeavour due to high shape variability. How can we efficiently and robustly represent highly variable anatomical shapes for image segmentation? We introduce three implicit shape representations, which are resolved using partial differential equations. Next, we compare their compactness and completeness using Principal Component Analysis (PCA) to summarise four classes of shapes. From a more effective combination of shape representation and PCA, we then derive a more stable and robust shape model. This shape model is integrated in image segmentation using the level set method, showing qualitative and quantitative results and improvements compared to previous methods.

Link of the Week: Uncovering Ancient Brews and Cures

When Patrick McGovern dons his royal purple latex gloves, the “Doctor Is In.” But this doctor isn’t working with live bodies; his “patients” are pottery shards from ancient China, Egypt, Lebanon and even Honduras.

Unlike traditional archaeologists who study the shards themselves for what they can tell us of past civilizations, McGovern, scientific director of the Biomolecular Archaeology Laboratory at the University of Pennsylvania Museum in Philadelphia, is looking for evidence of organic material in these remnants of jars, goblets and bowls.

McGovern and colleagues have identified the earliest alcoholic beverage in the world, dating back to about 7000 BCE, from a site in China’s Yellow River valley. His current focus is on what he calls “archaeological oncology” or “digging for drug discovery.” Read more.

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.