By Tim Stephens
WHEN IT WAS FOUNDED IN 1997, UCSC's Baskin School of Engineering had already established a reputation for excellence in computer science and computer engineering. But in the six years since its founding, the campus's first professional school has added a number of disciplines, emerging as a distinctive engineering school with a unique focus on some of the most exciting areas of technological innovation.
The faculty are conducting vital research in the core areas of information technology, biotechnology, and nanotechnology, and 12 of them were featured at the school's first Research Review Day in May (see "A Roundup of Engineering Research," below).
Graduates and students of the School of Engineering are also helping to build its reputation through their own accomplishments in industry and academia. In the pages that follow, we profile four of these people.
The subjects of the four profiles represent the school's founding departments, computer science and computer engineering. The quality of their work, however, speaks to the promise of the school's new departments and programs in applied math and statistics, biomolecular engineering, electrical engineering, information systems management, network engineering, and software engineering.
To accommodate the expansion of the engineering school's programs, a new building is under construction adjacent to the existing Baskin Engineering Building. With 90,000 square feet of new office, laboratory, and classroom space, the Engineering 2 Building is scheduled for completion in fall 2004.
Jack Baskin, whose $5 million gift helped launch the engineering school six years ago, has continued his support with a $1 million gift this year to help fund the new building and to create an endowed chair in biomolecular engineering. Baskin's donations to the School of Engineering now total almost $8 million.
"That we have come so far in such a short amount of time is due in large part to Jack Baskin's vision and support," says Chancellor M.R.C. Greenwood.
Kimmen Sjölander's specialty, bioinformatics, brings the methods of computer science to bear on problems in molecular biology. Her current work includes efforts to understand disease resistance in plants and, more broadly, the nature of innate immunity in both plants and animals.
The interdisciplinary nature of Sjölander's work is reflected in her affiliation with two departments at UC Berkeley: bioengineering and plant and microbial biology.
"It's easy for computer scientists to stay very theoretical, but by working closely with the biologists you find out what's really important to them," Sjölander says.
Proteins, with their extraordinary diversity of structure and function, pose some of the toughest problems in bioinformatics, and Sjölander has made key contributions to the arsenal of computational tools available for protein analysis. Her software programs can sort out the evolutionary relationships among proteins, allowing scientists to infer the structure and function of a newly discovered protein on the basis of its relationship to known proteins.
Sjölander began this work as an undergraduate working with bioinformatics pioneer David Haussler, UC Santa Cruz professor of computer science and Howard Hughes Medical Institute Investigator. After earning her Ph.D., also under Haussler's guidance, Sjölander worked in industry for several years. As chief scientist at the Molecular Applications Group (MAG), she oversaw the development of the Panther protein classification system, which included methods she had developed for her Ph.D. thesis.
When MAG's Panther group was acquired by Celera Genomics, Sjölander found herself working on the analysis of the human genome sequence. While Sjölander was at Celera, the public Human Genome Project recruited Haussler's group to help analyze its sequence. Although the media tended to focus on the competition between Celera and the public consortium, Sjölander says those involved just laughed about it.
"I think whatever competition there was got hyped up by the media," she says.
Nevertheless, she is thrilled to be back in academia, with greater freedom to do research in a collaborative environment. She was recently awarded a prestigious research grant from the National Science Foundation's Faculty Early Career Development (CAREER) program. And she was busy this summer organizing a research conference on bioinformatics held at Oxford University.
It is ironic that Sjölander ended up at UC Berkeley, having turned down generous offers from Berkeley in favor of UCSC for both undergraduate and graduate studies. The opportunity to study with Haussler was a big factor in those decisions, but there were other reasons as well, she says.
When Sjölander went back to school to earn her bachelor's degree, she was a single mother of three, including two-year-old twins. UCSC's Family Student Housing offered a supportive environment, including free after-school care for her children.
"The support that UCSC provided made a tremendous difference to my success," she says.
Aman Shaikh has not yet written his Ph.D. dissertation, but there can be little doubt about the quality of the work he has already completed. Shaikh has been working on his research at the AT&T Labs in New Jersey, where he helped develop a new software tool for monitoring network performance.
After extensive testing, AT&T has deployed Shaikh's monitoring software on its national backbone network, a core component of the Internet.
"As you can imagine, the barriers to such deployment are quite high, and getting to this point within such a short time is no small feat," says Shaikh's UCSC adviser, Anujan Varma, a professor of computer engineering.
Varma used his industry connections to create the opportunity for Shaikh to do his thesis research at AT&T, which has provided almost $100,000 in funding to UCSC for the project.
The stability and performance of a computer network depends on its routers, devices that manage network traffic and find the best routes for sending data packets to their destinations. Open Shortest Path First (OSPF) is a widely used routing protocol that enables routers to gather information about the network and find the best route to a destination. It does this by sending messages back and forth between routers.
Shaikh, working with AT&T researcher Albert Greenberg, developed an OSPF monitor that listens in on the constant chatter of messages (called Link State Advertisements, or LSAs) passing between routers. The monitor can detect problems in the network, such as a failing router, before they become serious.
"The OSPF monitor basically listens to LSA messages and analyzes them to assess the health of the network," Shaikh says."It does the analysis in real time, but it also archives the messages, so you can go back and do a more detailed analysis offline."
AT&T first tested the monitor on a small research network, then deployed it in a large customer network that connects hospitals to a health services data center. Shaikh's OSPF monitor detected a problem with one of the network's core routers, enabling the company to fix the problem before it affected customer service. That experience paved the way for the monitor's deployment in AT&T's national backbone network.
"Aman Shaikh's work has been extremely well received in the networking research community," says AT&T's Greenberg."It has been deployed in very large networks, where it has had significant real-world impact."
In addition to its practical value, the OSPF monitor is a useful research tool because of the data it collects, says Varma.
"We are getting a lot of information that we can use for subsequent research on network behavior," he says.
Shaikh continues to work on the project and plans to complete the requirements for his Ph.D. by the end of 2003.
Consumers in the United States have shown limited interest so far in navigation systems for cars, but Mike Tzamaloukas is developing a product that could change all that when it hits the market sometime next year.
Imagine having a device in your car that knows not only where you are and where you want to go, but also the current traffic conditions, and can map out the best route to get you around traffic jams and to your destination as quickly as possible.
That's the idea behind Circumnav, a company Tzamaloukas started incubating in 2002 with help from Skymoon Ventures, a venture capital firm based in Palo Alto. The company was officially founded this year.
"We've developed a wireless device that enhances the navigation systems now available for cars in a very affordable manner, and can give you dynamic route guidance based on traffic conditions," Tzamaloukas says.
Although Tzamaloukas is reluctant to reveal details of the technology behind his company's product, Circumnav uses the kind of wireless network technology that he has been working on since he was a graduate student with professor of computer engineering J. J. García-Luna-Aceves.
García-Luna's group developed technologies for establishing an ad hoc communications network consisting entirely of mobile wireless devices. It is like a network of cellular telephones that can communicate directly with each other, with no need for cell towers to relay signals.
Circumnav incorporates similar wireless networking technology in its navigation systems to provide constantly updated traffic information. For now, however, the source of the traffic information is a secret.
"We haven't been giving all the details, but we have the technology to make it work," Tzamaloukas says.
Before launching Circumnav, he was chief scientist with AmbiCom, a Fremont-based company specializing in wireless technology.
With the high-tech industry suffering during the recent economic downturn, this would seem like a difficult time to launch a new company in Silicon Valley. In the case of Circumnav, however, Tzamaloukas leveraged an idea that Skymoon Ventures was already exploring on its own. He now works out of the Skymoon offices in Palo Alto.
"They had a great idea, and I knew how to make it happen," he says.
Randal Burns is working to solve one of the great challenges of the digital age: how to manage efficiently the massive amounts of data stored in computer systems. It is a challenge that any large organization eventually confronts as data steadily accumulates and computer systems and networks evolve.
According to Burns, data storage accounts for most of the money spent on information technology and often represents the most valuable asset of a company or organization. A major focus of his research addresses the difficulty of moving data from one computer system to another.
"When data is stored in a certain system it tends to gain inertia, in the sense that it becomes harder to move it to a different system," he says."We're trying to overcome that inertia by developing tools that allow data to move between different software systems and different management domains."
These tools enable data to migrate seamlessly between computers running different operating systems and software programs. That can be helpful for sharing data among different sites and for deploying new storage technologies.
"We want to allow data to outlive the software systems on which it is initially stored," Burns says.
Burns directs the Hopkins Storage Systems Laboratory in the Whiting School of Engineering at Johns Hopkins University. As a graduate student, he worked with Darrell Long, professor of computer science and director of UCSC's Storage Systems Research Center. He also was on the research staff at the IBM Almaden Research Center in San Jose, both as a graduate student and after he earned his degree.
"One of the great things about my graduate career was the interaction with industry while I was at Santa Cruz," Burns says.
His work at IBM earned him six patents and a series of IBM awards for his inventions, including an IBM Outstanding Innovation Award. Since his arrival at Johns Hopkins in 2001, Burns has received two prestigious federal grants to support his research and teaching: an Early Career Principal Investigator Award from the Department of Energy and a Faculty Early Career Development (CAREER) Award from the National Science Foundation.
The NSF CAREER awards are meant to recognize and support those young faculty who are most likely to become the academic leaders of the 21st century. It's not hard to see why Burns was among those chosen.
A ROUNDUP OF ENGINEERING RESEARCH
THE BASKIN SCHOOL OF ENGINEERING
The Baskin School of Engineering held a Research Review Day this past May, presenting a sample of current research activities to an audience that included high-tech industry representatives from Silicon Valley. Some of UCSC's leading engineering faculty discussed their work in the series of presentations outlined here. Additional information about the talks described here is available on the web at soe.ucsc.edu/events/research_review_day.
Return to Fall 2003 Issue Contents