Interactive Classrooms over the Internet

ARSC and the University of Alaska Fairbanks (UAF) have joined forces with the Montana Rockies Center for Computational Science (MRoCCS) at the University of Montana (UM), and the Albuquerque High Performance Computing Center (AHPCC) at the University of New Mexico (UNM) to initiate the first for-credit high performance computing (HPC) course offered in three locations simultaneously via the Internet. The course is designed to introduce students with research interests in the physical sciences to the concepts of parallel scientific computation.

As scientists tackle more and more large-scale scientific problems, combined with the demand for higher performance in computer modeling and visualization,

A view of the screen shows the various groups participating in the computational science class. Students can see, hear and communicate with all participants shown on the screen. To the far right, a shared PowerPoint presentation supplements the discussion.

the need for research scientists to expand their familiarity with various advanced computational techniques has increased. However, if research scientists don’t have a background or familiarity with computational science, it’s not as likely they will take advantage of HPC as a tool to maximize their existing research programs.

“The idea is to treat computational skills as a necessary foundation equivalent to the field/lab skills scientists must possess to do quality research,” says Guy Robinson of ARSC. To prepare scientists to effectively use scalable parallel computing, the course focuses on the development of an intellectual and physical computing infrastructure to create a literacy among research scientists whose background doesn’t currently include computational methodology.

“You don’t need a supercomputer to learn parallel programming,” says Don Morton, of the Department of Computer Science at UM. “It’s less costly to train researchers in supercomputing methodology on small clusters of PC workstations. When they have mastered the training on PCs involving code development and debugging, and have developed a mature code, they can then easily run it on a supercomputer without incurring the high cost of a learning curve.”

The HPC course is offered at UAF, UM and UNM via the Access Grid (AG), an infrastructure consisting of a combination of hardware and software technologies linked together from various sites to support highly distributed collaborations in science, engineering and education. This infrastructure is rapidly developing into a worldwide information infrastructure linking supercomputers, virtual environments, scientific instruments, large databases and research teams.

The AG digitally links workgroup spaces called Access Grid Nodes (AGN), specifically equipped with high-end audio and visual technology, to provide a high-quality user experience. It supports large-scale distributed meetings, collaborative work sessions, seminars, lectures, tutorials and training. The focus is group-to-group communication—differing from desktop to desktop-based tools, which focus on individual PC communication.

UAF’s AGN, located in the Butrovich building, runs on a high-speed network. The technology includes large-format displays and multimedia projectors combined with high-quality multi-channel digital video and audio technology integrated with presentation technologies, such as distributed PowerPointTM and VNS, for slides and whiteboards.

Students in the UAF Computational Science class partcipate in a discussion with other class members over the Access Grid Node.

Each of the three classes has an individual instructor as well as a “node operator” on-site should technical difficulties arise with the AGN. Having instructors located at each site gives students the added benefit of expertise and feedback that each teacher can provide to the whole group.

At present there are nearly 50 AGNs across the United States, with approximately 20 more in other locations around the world. AGN technology has only been developed in the last few years and is still in the experimental phase. Most of the early AGN sites are members of the National Computational Science Alliance (NCSA), which is funded by the National Science Foundation’s (NSF) National Partnerships for Advanced Computational Infrastructure (NPACI).

Having an AGN gives UAF more opportunity to be a participant in the larger scientific community. The AGN can be a tool for isolated areas, like Alaska, Montana and New Mexico, to connect to the rest of the world without the need and expense of sending people elsewhere for training.

A State and National Resource…

The Arctic Region Supercomputing Center supports high performance computational research in science and engineering with an emphasis on high latitudes and the Arctic.

The center provides high performance computational, visualization, networking and data storage resources for researchers within the University of Alaska, other academic institutions, the Department of Defense and other government agencies. ARSC is located on the UAF main campus in Fairbanks, Alaska.

Arctic Region Supercomputing Center | PO Box 756020, Fairbanks, AK 99775 | voice: 907-450-8600 | email: