TeraGrid partners map the future of nation’s largest grid project

Partners in the National Science Foundation’s TeraGrid project held their quarterly meeting in Chapel Hill, NC,  March 13 – 15, where they discussed the impact of TeraGrid resources and infrastructure on scientific discovery and strategized on ways to involve the broader research community in the TeraGrid.  

Built and implemented over the last four years, the TeraGrid is the world’s most comprehensive distributed cyberinfrastructure for open scientific research. Through high performance network connections, the TeraGrid integrates high performance computers, data resources and tools, and high-end experimental facilities around the country.  The TeraGrid currently serves over 1,600 scientists and engineers who use computational, data management and visualization resources at eight sites across the U.S.

The Renaissance Computing Institute (RENCI, www.renci.org) hosted the meeting. RENCI Director Dan Reed helped develop the TeraGrid project in 2001 when he was director of the National Center for Supercomputing Applications. He coined the term TeraGrid to describe a grid infrastructure capable of moving and analyzing trillions of bytes of data – or terabytes. RENCI contributes to the TeraGrid through the Science Gateways initiative. Science Gateways work to make TeraGrid resources accessible to new  communities of users through common Web portal interfaces. RENCI works with three TeraGrid Science Gateway projects:

  • RENCI leads the Bioportal and Biomedicine Science Gateway project, which is developing a portal interface that will give biologists an easy-to-use interface for developing computational models, comparing large datasets including genomic sequences, accessing instruments and collaborating with colleagues. RENCI is adapting its North Carolina Bioportal for the TeraGrid.
  • The Linked Environments for Atmospheric Discovery (LEAD) Science Gateway is creating cyberenvironments for mesoscale atmospheric research. RENCI’s work with LEAD focuses on performance monitoring and adaptation and fault tolerance performability and recovery for the LEAD infrastructure.
  • RENCI participates in the Open Science Grid, a consortium of U.S. universities and laboratories that works to create a common national grid infrastructure that is open in its architecture, implementation and use.

More science, more successes

As the TeraGrid has matured, it has become an important resource for researchers working on a wide range of scientific problems. Examples include:

  • Harvey Newman, a particle physicist at the California Institute of Technology, ran simulations across multiple TeraGrid sites using Gridshell, which provides built-in support for job submission and scheduling, remote I/O redirection, parallel execution of commands/jobs, and inter-script task communication. Newman is investigating the discovery potential of CERN’s CMS experiment at the Large Hadron Collider. The work involves generating, simulating, reconstructing and analyzing tens of millions of proton-proton collisions.
  • Klaus Schulten, a University of Illinois molecular biologist, conducted simulations of the activities of nuclear pores, large protein complexes that allow water soluble molecules to pass into a cell’s nucleus but restrict other molecules from passing.
  • The MIMD Lattice Calculation (MILC) collaboration, led by Bob Sugar of the University of California at Santa Barbara, continues to be one of the largest users of TeraGrid resources. The group studies quantum chromodynamics, or QCD, which describes the span interactions that bind protons and neutrons together to form the nuclei of atoms.
  • In a TeraGrid use that extends beyond science, Rob Shakespeare, a professor in the theatre and drama department at Indiana University, is using TeraGrid resources to create virtual environments to replace traditional theatre sets.

The future: accessibility and interoperability

Grids, according to TeraGrid Director Charlie Catlett, can be much more than a system for linking and accessing powerful computing resources.

“There are growing data collections out there and researchers would be well served if they were integrated and easily accessible,” he said. “There is a need for collaborative services and for work flow systems that can readily incorporate TeraGrid resources. These are all important elements of the TeraGrid infrastructure and of cyberinfrastructure in general. We must extend these capabilities.”

The Science Gateway projects are an important ingredient in extending the capabilities and accessibility of the TeraGrid and new Science Gateway projects that operate seamlessly with the TeraGrid is a goal in the coming year, added Catlett.

In addition, TeraGrid technical staff will continue to implement tools that make collaboration, scheduling and other operations as easy as possible. These include advanced capabilities such as web services enabled by the Globus Toolkit (GT4), new scheduling services, expansion of the high performance, wide area distributed file systems and introduction of a TeraGrid user portal.

More information about TeraGrid facilities, science highlights, science gateways, and other programs is available at www.teragrid.org.

Relevant URLs:
TeraGrid Website
North Carolina Bioportal