Skip to main content

in this section

Welcome Immanuel Trummer

  • New Faculty (2016)

Before Immanuel Trummer even had a computer, he wrote programs on paper and gave them to his friends to try out on their computers. By the time he was in high school in Stuttgart he was writing and selling video games to his friends. Trummer grew up in a family of musicians and he began serious study of the violin at the age of five, winning several prizes at national music competitions in the following years. “When it came time to decide what to pursue in University,” says Trummer, “I had a tough choice. In the end I decided that I have more of a passion for Computer Science than for music, so that is what I chose to study.”

Trummer received a French-German Double Diploma in Computer Science and Engineering from the University of Stuttgart and the Ecole Centrale de Nantes. He then went on to earn a Ph.D. in Computer Science from the Ecole Polytechnique Federale de Lausanne (EPFL) in Switzerland. Trummer worked with Professor Christoph Koch at EPFL, where he began, among other directions, his study of various generalizations of the classical Query Optimization problem.

Query optimization is about figuring out how to process data in the most efficient way. Many popular languages for data access are nowadays declarative, meaning that users describe the data they want to generate instead of specifying how to generate it. Query optimization is the fundament for such languages as it shifts the burden of deciding how to execute a given query from the user to the computer.

“People have been working on the query optimization problem for more than 30 years,” says Trummer. “In my dissertation I argued however that the classical problem model is outdated. We have nowadays many more choices in data processing and having more choices makes the problem harder. In my dissertation, I show how to solve query optimization problems in seconds where prior methods need hours or days.” Trummer’s work has been well-received in the field. His papers have been selected among the best papers of the Very Large Data Base (VLDB) conference, for the Association for Computing Machinery’s (ACM) Special Interest Group on Management of Data’s (ACM SIGMOD) Research Highlight Award, and as Research Highlight in the Communications of the Association for Computing Machinery (CACM) journal.

Trummer joined the faculty of Cornell’s Department of Computer Science in July of 2016. “My main criteria for where I wanted to work are the quality of the faculty and the quality of the students,” says Trummer during a conversation in his office in the new Gates Hall. “I came to visit Cornell and I was very impressed.” Trummer is excited by the possibilities at Cornell. “Working with databases leaves many chances for collaboration,” says Trummer, “both through the people who collect the data sets and the people who create methods that can be used in data processing.”

Besides query optimization, Trummer now has multiple projects with several key research focuses. What ties his projects together is the overall goal of making data analysis more efficient and more user friendly. One idea Trummer is pursuing is how to create a tool that will allow users to interact with a database using natural language. In another project, he is working with Google on a tool that uses machine learning to text mine billions of subjective property associations from the Web. A third direction for Trummer’s current research is done in collaboration with NASA’s Ames Research Center. Together, they are exploring the potential for quantum computing to make data analysis more efficient in the long term.

“Most of what I do centers around the questions of how we can make data analysis more efficient and data access more user-friendly.” says Trummer. “There are nowadays more and more areas, be it in industry or in the sciences, in which you need large-scale data processing to make any progress. This development makes those questions more pressing than ever.”

Trummer is teaching a course in advanced database systems in the fall semester of 2016. In the spring semester of 2017 he will teach the introductory course in databases.

To see a talk of Trummer about his recent and ongoing research, click here:

back to listing