Database Pioneer Joins Microsoft to Start New Database Research Lab



David DeWitt, Microsoft Technical Fellow and director of the new Microsoft Jim Gray Systems Lab

REDMOND, Wash., April 23, 2008 — The database group of the Computer Science Department at the University of Wisconsin-Madison has been particularly prodigious in producing excellent talent in the database field in both industry and academia. Founded by, and until recently under the leadership of David DeWitt, the group has produced some of the top names in the field – including Microsoft Technical Fellows Peter Spiro and Rakesh Agrawal – and numerous others at Microsoft and throughout the industry at large.

DeWitt’s new role is creating and leading the Microsoft Jim Gray Systems Lab, a new advanced development center in Madison, Wisconsin for Microsoft, in association with the University of Wisconsin-Madison Computer Sciences Department. Peter Spiro, a former student of DeWitt’s and current Microsoft Technical Fellow, calls him “ … one of the top minds in database systems.”

“Attracting someone of David’s caliber is a coup for Microsoft,” says Spiro. “He has made exceptional contributions to the database field both through the research projects he has lead and the students he has educated over the past three decades.”

DeWitt is a member of the National Academy of Engineering, a Fellow of the American Academy of Arts and Sciences and former department chair and emeritus professor of computer science at the University of Wisconsin.

PressPass spoke with DeWitt about his new role and his vision for the new Microsoft Jim Gray Systems Lab.

PressPass: You’ve just joined Microsoft as a Technical Fellow. Tell us about your background and why you made the move to Microsoft.

DeWitt: I’ve just recently retired as an active Professor in the Computer Sciences Department and the founder of the database group at the University of Wisconsin-Madison. I taught at the university for the past 32 years and thoroughly enjoyed my teaching and publishing opportunities there.

I’ve long been passionate about the database field. I’ve been privileged to work with groups that have pioneered the areas of parallel databases, database system performance evaluation, object-oriented database systems and XML databases. My research has also focused on advancing the state of the art of database systems through the design, implementation and evaluation of working systems. I believe that, in the future, databases systems can play a very important role in making research in domains like medicine and large scale science more productive.

I also believe that by applying database technologies to the management of very large datasets, such as those that are obtained by web-crawling, we can dramatically reduce the cost of analyzing them. Applying database technologies has the potential for reducing the cost of analysis by at least a factor of 10. These types of theories are what the lab will be exploring.

PressPass: What will the Microsoft Jim Gray Systems Lab be working on?

DeWitt: The ability to give users real-time access to the most up-to-date data – any type of data – is the core of Microsoft’s vision. We’re at a pivotal point in the database industry, and our lab will be at the forefront of this rapidly changing environment. Data is the most critical organizational asset. The quantity and variety of data within most organizations are rapidly increasing, thanks to new technologies like digital images and video and RFID.

The traditional definition of data itself is also changing—databases today go way beyond facts and figures to store new data types such as x-rays, videos and even spatial content. And there are a variety of options for delivering this data, from mobile phones to laptops to our traditional desktop and server environments. Microsoft’s data platform is leading the evolution of the database to manage new data types, support real-time access, and provide pervasive insight for business decision makers.

In recent years, funding for core database systems research from the federal government has become very limited. This lab will make it possible for graduate students at the University to continue exploring fundamental database system issues. Having access to SQL Server source and the “cloud” computing facilities at Microsoft will also enable the graduate students at Wisconsin to explore unique avenues of research that are not open to graduate students elsewhere.

More specifically, the Lab will perform advanced development on Microsoft codebases to generate ideas about future directions for data management. We will be developing these ideas to the point where they can be implemented. The Lab will work closely with Microsoft researchers in Redmond and other locations. The Lab will follow a flexible model of innovation and incubation in which the Lab’s staff will coordinate its work with Microsoft’s core development teams, to ensure that we maximize the opportunity for learning and future integration.

PressPass: Why is the lab being located in Wisconsin?

DeWitt: By being a part of the systems and database community in Madison, the lab will be able to contribute to and feed off of that community’s energy and thought leadership. We also intend to draw on the faculty of the University of Wisconsin-Madison Computer Sciences Department as consultants.

PressPass: Why did Microsoft decide to name the facility the Microsoft Jim Gray Systems Lab?

DeWitt: This is an emotionally charged question for me. Jim was a very close friend and my mentor from very early on in my career as a beginning assistant professor. No one assigned him this role. It was a role he took on for me and countless other junior faculty members over his lifetime. Jim was also very interested and supportive of education throughout his career and became a strong advocate of computer sciences at the University of Wisconsin-Madison. Having his name on the lab is a way of honoring his family, his technical contributions to the field as a Turing award winner, his support of education and research, and his relationship with the department. His contributions were countless.