DAILY NEWS AND INFORMATION FOR THE GLOBAL GRID COMMUNITY / JULY 21, 2003; VOL. 2 NO. 29    ( Table of Contents )

Special Features:

APPLICATION TRENDS IN VERTICAL MARKETS, PART I
by Insight Research

Grid computing technology was cultivated within the research and academic communities as a way for scientists to address the rapidly increasing needs for computing power, data access across wide areas, and virtual-community collaboration. Commercial enterprises, however, are increasingly deploying grid computing technologies in both mundane and innovative ways. For example:

Commercial Applications of Grid Computing

Some of these scenarios are being executed today, and others are quickly approaching feasibility. Grid computing technology is currently being used in three broad areas:

• government and academic applications;
• commercial enterprise applications; and
• consumer applications.

INSIGHT Research has elected to use a model consistent with these three categories for the phased introduction of grid applications over the next few years, and as shown in the table below.

Phased Introduction of Grid Applications

Government and Academic Applications

To date, government funding is the prime driver of most research efforts in Grid Computing. Around the world, nations and groups of nations understand the development of grid computing to be a critical element of their evolving national cyber-infrastructures. In the US, this is evidenced by significant investments in the TeraGrid, and numerous other projects executed by the NSF and other agencies. In the UK, expectations for grid computing are demonstrated by the e-Science project and the European Union (EU) DataGrid, which represents a significant European community-wide effort. Similar efforts are underway in other countries around the world.

Although the traditional driver for scientific and research grids has been the need for high-performance compute power at a low cost, to many organizations, equally critical is the need to collect information from a variety of widely dispersed sources, and to share information among widely dispersed and different users.

Search for Extraterrestrial Intelligence

The well-known SETI@home project, managed by a group of researchers at the Space Sciences Laboratory of the University of California, Berkeley and supported by a number of private, corporate, and not-for-profit sponsors, was the first attempt to use large-scale distributed computing to perform a sensitive search for radio signals from extraterrestrial civilizations. Since 1999, Search for Extraterrestrial Intelligence (SETI) has used the spare capacity of more than four million PCs to analyze signals from outer space for signs of extraterrestrial intelligence. All PC cycles were contributed by volunteers from over 225 countries, while half of these cycles were attributed to the US. In the context of grid computing, SETI@home can be considered a rudimentary version of a public-service compute grid.

As of December 2002, the SETI@home project logged a total CPU time of 1,247,830 years and 2.321 Flops, the largest computation on record. So far, none of the signals examined has shown verifiable evidence of extraterrestrial intelligence. The project, however, has been considered a success, and several follow-up projects including SETI@home II are in the planning stages.

Earthquake Engineering Simulation

The Network for Earthquake Engineering Simulation (NEES) project, sponsored by the NSF, will link earthquake researchers across the US with leading-edge computing resources and research equipment, allowing collaborative teams to plan, perform, and publish their experiments. Called NEESGrid, this project will create a national virtual laboratory for earthquake engineering research and simulation. A major objective of the project is to shift the emphasis of earthquake engineering research from current reliance on physical testing to integrated experimentation, computation, theory, databases, and model-based simulation.

The initial three sites became active in September 2002 at the University of Nevada in Reno, Oregon State University in Corvallis, and Rensselaer Polytechnic Institute in Troy, NY. By September 2004, the Grid is planned to support virtual collaborations among a minimum of 15 earthquake engineering sites around the country.

High Energy Particle Physics and Earth Observation Applications

The EU DataGrid is a project funded by the EU. Once completed, it will be one of the largest academic Grids in the world. Its objective is to build a computing infrastructure to enable the next generation of scientific exploration, which requires intensive computation and analysis of shared large-scale databases, from hundreds of Terabytes to Petabytes, across widely distributed scientific communities. The DataGrid will enable access to geographically-distributed computing power and storage facilities belonging to different institutions. This will provide the resources necessary to process the huge amounts of data coming from European scientific efforts.

The three data-intensive computing applications areas covered by the project are:

• high-energy physics,
• biology and medical image processing, and
• Earth observations.

UK e-Science Program

In the UK, governments are investing over US$ 200 million in the UK e-Science project. The mission of this project is to ensure that British industrial and research organizations have the necessary infrastructure as science migrates to e-science.

Of the more-than-twenty anticipated scientific projects that will be funded as part of the overall e-science initiative, about half are underway, including:

• Molecular modeling of condensed matter
• Combinatorial chemistry
• Distributed aircraft maintenance
• Engineering design optimization
• High throughput informatics
• Particle physics prototype grid
• Virtual observatory
• Oceanographic diagnostics and visualization
• Global climate research
• Environment modeling at the molecular level
• Aerospace applied research

Cancer Diagnosis and Screening

The University of Pennsylvania has received grants over several years from a number of US federal government agencies to develop a next-generation Electronic Medical Record data grid and repository. The vision is to develop a patient-centric medical record system that could capture the full range of healthcare files, including high-fidelity patient medical images, records, and clinical history-all from any location.

In a similar project, Oxford University recently joined with a major IT service provider and the UK Government to build a sophisticated compute grid that will enable early screening and diagnosis of breast cancer, and provide medical professionals with more information to help treat the disease. This compute grid will make it easier for doctors to access patient mammograms and compare them with millions of others. The project is part of the UK government's eScience initiative.

High Resolution Neurosciences Imaging

Funded by the National Institute of Health, the Biomedical Informatics Research Network project is intended to address the increasing needs of biomedical researchers to access and analyze huge amounts of data at various levels of aggregation, which are located at academic and research sites throughout the country. The success of the project depends on the new computational and networking technologies that have been developed to bring researchers together over the Internet, on still newer technologies for federating data from multiple sources, and even more on the social integration of the scientific groups that will be sharing their data and, ultimately, making their collective results public. According to the project's Web site "BIRN is a test of a new mode of doing large-scale medical science."