Undergraduate Research

In Computer Science and Engineering

Why wait till grad school to be a star?

Department of Computer Science and Engineering

College of Engineering and Information Technology

University of South Carolina

Columbia,South Carolina

Students in the Computer Science and Engineering Department at USC aren’t just going to classes these days. An increasing number of them are doing things that used to be done only by graduate students and professors. They’re doing research—branching out on their own to discover new areas of knowledge, having a lot of fun in the process, and getting national and international recognition for their work..

Stacey Ivol and Matt Elder, for example, spent this past summer at an NSF-funded program in Florida doing robotic control. Stacey is from Pennsylvania; Matt is from Lexington, South Carolina. Their research paper, “Extensions of a Cooperative Control System,” is being submitted for publication to a professional journal. Cooperative control involves a group of robots collectively performing a common function or task. Cooperative control would allow a swarm of inexpensive robots to perform hazardous tasks that might be difficult or expensive for a single, complicated machine to do.

Another student, A.J. Alon, from Columbia, spent his summer doing research as an undergraduate intern at NASA’s John Glenn Research Center in Cleveland. The lab in which A.J. worked specializes in developing new technology to perform non-invasive measurement of the human body, with the intent of detecting diseases from glaucoma and macular degeneration to Alzheimer’s much earlier than before. A.J.’s summer research focused on showing that tools that presently monitor the blood flow in the eye can also be used to measure the much lesser blood flow in the fingertips. With luck, further research will allow this technology to measure fingertip hemodynamics during an astronaut’s walk in space. A.J. is being flown to Japan this fall by NASA to present his work at the International Astronautics Conference.

Joe Turner, from Pawley’s Island, and Chris Jones, from Sumter, South Carolina, have been working in the USC lab for scalable parallel computing on problems in high performance computing. Joe’s work has been in performance analysis, with a focus on the

use of on-chip performance counters, using the monitoring capability of the 64-bit Intel Itanium processor to develop performance prediction metrics as problems grow in size and as the number of processors increases. Using the 256-node SGI Altix at Oak Ridge National Lab, Joe is increasing the performance of an ocean-simulating code. He will soon be finished with a paper to submit to a professional conference.

Chris Jones is observing the energy consumption and time-to-solution for the High Performance Computing Challenge benchmark problems. He is constructing power profiles and searching for slack time in the benchmark codes, when the processor is not fully utilized, so that in real time the voltage on the CPU could be lowered and the processor frequency scaled back to save power. Saving energy in a high performance parallel computer can be very important—some of the Department of Energy’s computers have literally thousands of processors and consume power measured in megawatts. His work is being submitted for publication.

These current students in computing from USC join several recent graduates in garnering national recognition. Blaine Nelson, now in the doctoral program at Cal-Berkeley, was a Computing Research Association Undergraduate of the Year Honorable Mention in the 2003 class. Heather Wake, class of 2004, was the CRA female Runner-up, won a prestigious National Science Foundation graduate fellowship (and three other fellowships as well), and is now in the doctoral program at Duke University. Her undergraduate Honors College thesis led to a conference presentation in California and a book chapter. Achraf El-Allali, in the 2005 class, was another CRA Honorable Mention, and is also now in graduate school.

Heather Wake at FCCM in Napa, CA

For more information, contact

Department of Computer Science and Engineering

University of South Carolina

Columbia, SC 29208

803-777-2880

www.cse.sc.edu