With a nod to the newly formed Virtual Institute - High Productivity Supercomputing (VI-HPS), we thought it might be useful to provide a link to the Keynote talk given at the institute’s inauguration a couple of weeks ago. Doug Post, the chief scientist for the DoD High Performance Computing Modernization Program (HPCMP), gave a very informative talk about current/future challenges and opportunities in computational science and engineering. Loaded with statistics and information about challenges and bottlenecks related to code development and deployment, most of the talk is centered around HPCMP work, but much of it is applicable to HPC efforts within the US national research agenda.

What is a virtual institute you ask? In the VI-HPS context, a virtual institute is a specific research entity that unites distributed competencies and is funded by Germany’s Helmholtz Association.

The President’s NSF budget request for 2007 contains some good news. The NSF Office of Cyberinfrastructure is slated for a $55 million increase and cybinfrastructure R&D is slated for a 15% increase over the previous budget, which sends a strong message, especially considering the overall budget increase is only 7.9%. Visit NSF’s official budget site for full details or a summary can be found here. Also check out the Computing Research Policy Blog for a brief analysis of the budget.

The budget is good but how much of it Congress cuts remains to be seen.

As disciplines such as biology, ecology, chemistry, etc. become more ensconced with computational science, information needs and sharing increase. One such need is the demand for information professionals, especially academic librarians, who provide invaluable services for students and researchers. The University of Tennessee is one institution addressing this need, becoming a recent recipient of a grant to educate and train next generation science information professionals and strengthen the information infrastructure on which research is founded.

The Australian Research Council (ARC), an Australian equivalent of the NSF, recently awarded more than $3.5 milliion over the next couple of years for grid computing technologies aimed to increase medical research collaboration. One key beneficiary of the grant, Dr. Andrew Lonie of the University of Melbourne, will be using his share of the funds to work on the international Physiome Project, the successor to the Human Genome effort, which has a goal to

describe the human organism quantitatively, so that one can understand its physiology and pathophysiology, and to use this understanding to improve human health.

As part of this new ongoing effort, Dr. Lonie’s research centers around modeling and simulation of the human kidney, via the Kidney Simulation Project.

Continued funding for grid technologies and the maturation of high-speed networking will boost opportunities for international reearch collaboration and engagement. The result will be the ability to link the worlds foremost authorities in medical science to massive amounts of data, which will ultimately lead to quicker solutions to, and better treatment for, both local and global health issues.

Information Week covered the new TeraGrid dollars in its most recent issue. Scientific Gateways get play:

Historically, scientists who needed access to the most powerful computers have been willing to adapt their work to the requirements of large supercomputers, Catlett says. The goal of the science gateways is to provide Web applications and PC software that can give scientists in a given field a common way of running programs on Teragrid machines. Says Catlett, “It’s one of the most exciting things we’re doing.”

NSF recently awarded a group of universities $10 million over five years to set up and operate a grid that will allow researchers and students to access physics data produced by the Large Hadron Collider at CERN in Geneva, Switzerland. The Data Intensive Science University Network, or DISUN for short, will provide access to results from the Compact Muon Solenoid (CMS) experiment, which will account for a portion of the petabytes of data produced by the Collider annually. The CMS effort will also contribute to other grid projects including the Open Science Grid.

More detailed information about the project can be found in Supercomputing Online’s story about DISUN from last week.