Posted April 7th, 2014
Google has granted a Faculty Research Award to Professor Torsten Suel to support his study of new index pruning and index tiering techniques that could significantly reduce hardware and energy costs for large web search engines. “Google receives 5 billion queries a day, and the company indexes trillions of pages. This huge data size and query load mean that running a large, state-of-the-art search engine is now a very expensive undertaking, requiring massive amounts of energy,” explained Suel, a professor in the Department of Computer Science and Engineering and an expert on search engine architecture.
Google pulls its search results from an index—think of a massive digital version of the index in the back of a book. Suel’s aim is to prune that index by finding ways to eliminate word occurrences that are unlikely to be helpful during a given search. Additionally, he is studying index tiering systems, in which parts of the index that are considered less useful are consulted only on selected queries.
This is the third Google award given to Professor Suel, who holds a Diplom degree from the Technical University of Braunschweig and a doctorate from the University of Texas at Austin. Google, which estimates that its index is now well over 100 million gigabytes, bestows the Faculty Research Award, as the unrestricted grant is known, upon scientists working in areas of key interest to the company as well as to the broader research community. Suel’s grant is for $55,500.