image placeholder

2012 Lancaster Dissertation Award in Mathematics, Physical Sciences, & Engineering

The Department of Computer Science is delighted to announce that Dr. Sudipto Das has been awarded the 2012 Lancaster Dissertation Award in Mathematics, Physical Sciences, & Engineering. This is the first time this prestigious and highly competitive campus-wide award has been awarded to a Computer Science PhD student. Sudipto Das’s PhD dissertation is entitled “Scalable and Elastic Transactional Data Stores for Cloud Computing Platforms,” was conducted under the supervision of Professors Divyakant Agrawal and Amr El Abbadi. He graduated in December 2012, when he joined Microsoft Research (MSR).

The award is given by the UCSB Graduate Division to one doctoral dissertation completed in Mathematics, Physical Sciences, & Engineering in the two year period of July 1, 2010 to June 30, 2012. The departments for this year’s competition included Chemistry, Chemical Engineering, Computer Science, Earth Science, Electrical and Computer Engineering, Environmental Science and Management, Mathematics, Material Science, Mechanical Engineering, Media Arts and Technology, Physics, and Statistics. The winner is selected based on the impact on field in terms of methodological and substantive contributions.

Sudipto’s dissertation targets the important problem of designing scalable and elastic database management systems (DBMSs) for storing and serving large amounts of data (referred to as “big data” in common folklore). DBMSs form a critical component in the success of a vast majority of enterprise and web-applications that are part of our daily life. As the cloud computing platforms become an integral component of our daily interactions with computers, scalable and elastic DBMSs serving these applications are critical.

Sudipto’s dissertation makes several fundamental contributions towards realizing the vision of building scalable and elastic OLTP DBMSs for cloud platforms. These contributions significantly advance the state-of-the-art by supporting scale-out transaction processing and promoting elasticity as a first class concept in database systems. The significant technical contributions are in schema-defined and dynamically-specified database partitioning, large self-managing DBMSs installations, virtualization in the database tier, and live database migration for elastic load balancing. These technologies are critical to ensure the success of the next generation of DBMSs in cloud computing infrastructures.