September 23, 2003

Charles Elkan, CSE Associate Professor and a member of the Interfaces and Software Systems layer of the UCSD Division of Cal-(IT)2, has received a $50K donation from Sun Microsystems to support research on intelligent data mining. The funds will be used to support a second- or third-year Ph.D. student possibly to extend work on clustering algorithms conducted by recent Ph.D. graduate Greg Hamerly.

Clustering algorithms analyze data sets looking for homogenous subsets of data. Potential applications of this work include new classifications of protein structures into families and anomaly detection to enable further investigation of aberrant behavior in machinery before an actual failure appears.

Says Elkan, "Typically, faculty are hired by companies as consultants. A partnership of this kind is much more unusual - and exciting - because it enables me to learn about real-world problems and it gives me the opportunity to do basic research."

Elkan is working with Kenny Gross, Senior Physicist, Sun Microsystems. Gross is based in San Diego at Sun's San Diego Physical Sciences Center, which focuses on further improving the reliability and availability of high-end (>$1M) Sun systems. In the Physical Sciences Center, Sun is applying the sensornet concept to monitor internal system performance to provide predictive warnings of potential system problems so that outages can be prevented.

Says Gross, "Technology research collaborations between Sun and a world-class research institution such as UCSD help narrow technology gaps in next-generation computer servers. The partnership provides UCSD graduate students access to state-of-the-art technology at the Physical Sciences Center for their thesis research. We expect this partnership with Prof. Elkan to yield new advances in proactive health monitoring of enterprise-class computing servers."

Story Adapted from Cal-(IT)2 Press Release
Full Story can be found in the Cal-(IT)2 News Room
Media Contact: Doug Ramsey, (858) 822-5825