• June 2010
    M T W T F S S

TCD Researchers Collaborate on International Project to Integrate ‘Cloud Computing’ With ‘Grid’ Technologies.

Trinity College Dublin (06/22/10)

European researchers working on the StratusLab project are developing software designed to improve distributed computing infrastructures in an effort to enable research and higher education organizations to pool computing resources. Trinity College Dublin (TCD) researchers are leading the development of a repository of virtual appliances that will make it easier to create grid systems. The virtual appliances repository is designed to facilitate the growth and availability of grid computing for researchers. “The StratusLab toolkit will make the grid easier to manage and will allow grids to tap into commercial cloud services to meet peak demands,” says TCD’s David O’Callaghan. “Later it will allow organizations that already provide a grid service to offer a cloud service to academic users, whilst retaining the many benefits of the grid approach.” The StratusLab project enhances the distributed computing infrastructure ecosystem by simplifying management, adding flexibility, and increasing maintainability, quality, energy efficiency, and resiliency of computing sites.


Remembering Rajeev Motwani..

Rajeev Motwani (Hindi: राजीव मोटवानी; March 26, 1962 – June 5, 2009) was a professor of Computer Science at Stanford University whose research focused on theoretical computer science. He was an early advisor and supporter of companies including Google and PayPal, and a special advisor to Sequoia Capital. He was a winner of the Gödel Prize in 2001.

-Remembering Rajeev-

Sandia to Play Major Role in DOE-Funded Simulation of “Virtual” Nuclear Reactor.

Sandia National Laboratories (06/14/10) Singer, Neal

Sandia National Laboratories computational scientists are leading a U.S. Department of Energy effort to create a “virtual” nuclear reactor that will be headquartered at Oak Ridge National Laboratory’s Nuclear Energy Modeling and Simulation Energy Innovation Hub. The virtual reactor will use supercomputers to attempt significant leaps forward in nuclear reactor design, engineering, and operation. The virtual reactor should enable engineers to improve reactor safety, increase reactor power production, and extend reactor life. The Hub, named the Consortium for Advanced Simulation of Light Water Reactors (CASL), includes partners from universities, industry, and other national laboratories. Sandia scientists also will work with researchers from North Carolina State University to incorporate advanced methods for verification and validation of the computational models and for quantification of uncertainties as they are executed within the virtual reactor. “This Hub has the ability to transform the nuclear power industry through the research, development, and application of new mathematical models, advanced computational methods, and leading-edge high performance computing,” says CASL’s Jim Stewart.


Europe’s Prace HPC Grid Aims for Exaflop Power by 2019.

ZDNet UK (06/10/10) Meyer, David

The European Commission (EC) is funding a project by the Partnership for Advanced Computing in Europe (Prace) to develop a grid of up to six “tier 0” high-performance computing systems spread across participating member states. Prace is aiming for a combined computing power in the multi-petaflop range in the next five years, and in the exaflop range by 2019. “I warmly welcome the launch of the Prace supercomputer infrastructure as scientific computing is a key driver for the development of modern science and technology and for addressing the major challenges of our time like climate change, energy saving, and the aging population,” says EC Digital Agenda commissioner Neelie Kroes. Prace will help investigate photosynthesis at a sub-atomic level, which could accelerate the development of more efficient solar cells. “Scientists can also investigate [three-dimensional] protein folding, which helps them understand how drugs interact with cells in the human body,” according to the EC.


Intel, IMEC and Five Flemish Universities Open Flanders ExaScience Lab.

IMEC (06/08/10) Weingartner, Markus

IMEC, Intel, and five Flemish universities have opened the Flanders ExaScience Lab at IMEC’s research facility in Leuven, Belgium. The lab will develop software to run on Intel-based exascale computer systems. The ExaScience Lab will become part of Intel’s European research network, which consists of 21 labs employing more than 900 researchers. The Flanders ExaScience Lab will focus on enabling scientific applications, beginning with the simulation and prediction of electromagnetic activity in the space surrounding Earth’s atmosphere. “We are excited to embark on this innovative collaboration in Flanders, bringing together scientific expertise in the fields of space weather prediction, computational simulation, reliability, visualization, and performance modeling,” says Intel’s Stephen Pawlowski. The Flanders ExaScience Lab will include researchers from the University of Antwerp, Ghent University, Hasselt University, Katholieke Universiteit Leuven, and Vrije Universiteit Brussel.


Seven Atom Transistor Sets the Pace for Future PCs.

BBC News (05/24/10)

A working transistor that contains only seven atoms has been built by a team in Australia. The researchers, led by University of New South Wales professor Michelle Simmons, developed the atomic transistor as part of a project to create a quantum computer. The atoms in a silicon crystal were replaced with phosphorus atoms using a scanning tunneling microscope. “Now we have just demonstrated the world’s first electronic device in silicon systematically created on the scale of individual atoms,” Simmons says. The working transistor was handmade, so there is a need for a process for developing them in large numbers. The researchers say the device could lead to chips that have components that are up to 100 times smaller than those on current processors. Simmons says the development could result in an “exponential” leap in processing power.


Researchers Race to Produce 3D Models of BP Oil Spill.

Computerworld (05/26/10) Thibodeau, Patrick.

The U.S. National Science Foundation recently made available an emergency allocation of 1 million compute hours on the Texas Advanced Computing Center’s Ranger supercomputer to study how the BP oil spill will affect coastlines. The goal is to produce a three-dimensional (3D) computer model that can forecast how the oil may spread in environmentally sensitive areas by showing in detail what happens when it interacts with marshes, vegetation, and currents. The model “has the potential to advise and undergird many emergency management decisions that may be made along the way, particularly if a hurricane comes through the area,” says University of North Carolina professor Rick Luettich. The model, called Advanced Circulation Model for Oceanic, Coastal and Estuarine Waters, can track the oil spill into the marshes and wetlands due to its fine scale resolution, says University of Texas professor Clint Dawson. The 3D modeling can show what happens to the oil at various depths and how it travels as it comes in contact with underwater surfaces.


Social Supercomputing Is Now.

ETH Zurich (05/28/10)

ETH Zurich scientists working on the FuturIcT project plan to use the world’s largest supercomputers to simulate life on Earth, including the financial system, economies, and whole societies. The ETH researchers, working under the Competence Center for Coping with Crises in Complex Socio-Economic Systems, are analyzing huge amounts of financial data to detect dangerous bubbles in stock and housing markets, potential bankruptcy cascades in networks of companies, or similar vulnerabilities in other complex networks such as communication networks or the Internet. The FutureIcT project aims to bring different kinds of research together to simulate the entire globe, including the diverse interactions of social systems and of the economy with the environment. The FutureIcT project also aims to analyze data on social, economic, and environmental processes by augmenting the results of field studies and laboratory experiments. “Such observatories would detect advance warning signs of many different kinds of emerging problems, including large-scale congestion, financial instabilities, the spreading of diseases, environmental change, resource shortages, and social conflicts,” says ETH’s Dirk Helbing.


Customizing Supercomputers From the Ground Up.

Pacific Northwest National Laboratory (05/26/10) Beckman, Mary

The Pacific Northwest National Laboratory (PNNL) computer scientist Adolfy Hoisie will lead a group of scientists that will design supercomputers and their software applications simultaneously, so all the components of a supercomputer can be optimized and focused on one kind of problem. The PNNL team plans on solving the kinds of problems specific to various scientific fields, from studying biological systems to understanding the electrical power grid. The data-intensive problems the PNNL team wants to solve requires a different emphasis in computational resources, which is why they will design supercomputers and their applications simultaneously. “The science of systems and applications designed for optimal performance is a grand challenge for high performance computing research,” says PNNL researcher Moe Khaleel. The team also will examine how performance and power relate, and how they trade off against one another on extreme-scale systems and workloads.


Tokyo Tech Announces Plans for 2.4 Petaflop Supercomputer.

Tokyo Institute of Technology (05/26/10)

The Tokyo Institute of Technology (TIT) announced that the TSUBAME 2.0 supercomputer, a green, cloud-based supercomputer system with a top speed of 2.4 petaflops, will begin operation this fall. TSUBAME 2.0 will be built by Hewlett-Packard and NEC using GPGPU computing and will feature a large solid-state drive. TSUBAME 2.0, dubbed Petakon, will be 12 times faster than Japan’s current fastest supercomputer and is expected to achieve a top ranking on the TOP500 list. It also is expected to score high on the DARPA HPC Challenge benchmark and the Green 500 list. TSUBAME 2.0 will feature Intel Westmere-EP and Nehalem-EX processors with scalar operation, and will employ approximately 4,200 NVIDIA Fermi graphics processing units. Petakon also will have more than 1,400 computer nodes and use Voltaire’s QDR InfiniBand network. The operating system will be a combination of Linux and Microsoft Windows HPC, and will use virtual machine technology to provide cloud-hosting services.