Samuel Madden ’99, MNG ’99, MIT College of Computing Distinguished Professor of Computing, CSAIL member, and Cambridge Mobile Telematics co-founder, recently earned the Edgar F. Codd Innovations Award from ACM’s Special Interest Group on Management of Data (SIGMOD) for his contributions to the data management field. He received SIGMOD’s highest honor in recognition of his research in "column-oriented database systems, high-performance transaction processing, and systems for mobile and sensor data."
The Edgar F. Codd Innovations Award is given annually to a researcher who has made significant advances in the field of databases and database systems. It was first awarded in 1992 to MIT EECS adjunct professor and CSAIL member Michael Stonebraker.
As the most recent recipient of this prize, Madden currently leads the Data Systems Group at MIT and the Data Science and AI Lab (DSAIL). He develops new systems, algorithms, and interfaces to help people query and analyze data, ranging from sensor readings to massive relational tables to imagery, video, and text.
One area where Madden has worked extensively is the design of novel relational database systems, including column-oriented databases for analytics and database systems for high-performance transaction processing. For example, in the prototype C-Store, Madden and other collaborators from CSAIL, including CSAIL PI Michael Stonebraker, showed that storing tables as a collection of columns, instead of row-by-row, leads to more efficient queries, especially when performing complex analytics, statistics, and machine learning on large data sets with many columns. Madden's contributions in this space include building query executors that directly operate on compressed data and developing novel query processing strategies that focus on column-at-a-time execution of database operations.
Madden also developed the CarTel system with MIT professor Hari Balakrishnan, starting in 2005. In CarTel, Madden explored issues related to the collection of data from mobile devices, particularly vehicles, and developed several applications where sensor-equipped cars are used to measure the world, mapping traffic, wireless networks, and potholes. Based on the fundamental technology developed in the CarTel project, Madden and Balakrishnan founded Cambridge Mobile Telematics (CMT) in 2010, which provides technology to help make roads safer using a variety of technologies, including mobile sensing, machine learning, and behavioral science. As the world's largest telematics provider, CMT's technology is used by millions of drivers every day in over 18 counties.
More recently, Madden’s research has shown that the algorithms and other data structures in database systems can employ machine learning for increased efficiency, and on developing novel data processing systems that employ machine learning to process data. For example, in his DataHub project (commercialized as Instabase), Madden and his students focused on the extraction of tabular data from unstructured documents such as text files, spreadsheets, forms, and the like. In another series of projects, Madden and his students have built systems to provide SQL-like query interfaces over large archives of video, and more recently has been focused on using large-language models to provide database-like querying capabilities over large collections of text.
Madden received his PhD at the University of California, Berkeley in 2003, then joined MIT and CSAIL in 2004. During his time at Berkeley, he developed TinyDB, a lightweight data management system for networks of sensor devices. This project allowed users to issue SQL-like queries to networks of devices in physical spaces, for example to query the temperature on different floors of a building, the humidity throughout a redwood grove, or the occupancy of bird nests on remote islands. TinyDB pioneered several fundamental technologies, most notably the idea of "in-network aggregation," where database queries to reduce data are executed inside a network of devices, rather than being processed centrally.
Before this award, Madden was named one of Technology Review's Top 35 Under 35 in 2005 and an ACM Fellow in 2020. He’s also earned an NSF CAREER award, a Sloan Foundation Fellowship, and "test of time" awards from VLDB, SIGMOD, SIGMOBILE, and SenSys. In addition to CMT, his industrial impact includes co-founding Vertica Inc., which was bought by HP in 2011, and advisory positions with Heavy.AI, TileDB, and Instabase.
