Srini Devadas is the Webster Professor of Electrical Engineering and Computer Science and has been on the MIT EECS faculty since 1988. He served as Associate Head of the Department of Electrical Engineering and Computer Science, with the responsibility for Computer Science, from 2005 to 2011. His research focuses on Computer Aided Design (CAD), computer security and computer architecture. In 2015, he received various awards including: ACM/IEEE A. Richard Newton Technical Impact award in Electronic Design Automation, IEEE Computer Society Technical Achievement Award (2014) for inventing Physical Unclonable Functions and single-chip secure processor architectures.
Industry Impact
Protection of information has been a major challenge since the start of the computer age. Given the widespread adoption of computer technology for business operations, the problem of protecting information has become a more crucial epidemic. Classified computer files, databases, and Internet applications are the most sole and vital asset of an organization. Hence, when these assets become attack, threatened, or damage data integrity becomes questioned and conducting business is at a halt.
Recent works
Secure Demand Paging for Trusted Execution Environments
Trusted execution environments allow the processor to run a program without requiring trust in any software. This makes enclaves particularly relevant in cloud environments where there is no control of the software running on the machine. However, some indirect attack vectors known as side channels observe information about program execution such as what order pages are accessed or which cache lines are read. This project aims on altering the way trusted execution environments handle paging to decouple page access patterns from the control flow of the program, preventing an attacker from using these access patterns to learn information about program secrets.
Scaling Strong Anonymity
In an era of mass surveillance, maintain anonymity on the Internet is an important yet very difficult challenge. “Tor”, the only widely deployed anonymity system, unfortunately fails to provide anonymity against an adversary that can globally monitor the Internet. However, most provably secure anonymity systems fail to scale to a large number of users, preventing wide adoption of such systems. The goal of this project is to create systems that can provide both strong anonymity while scaling to millions or more users. In efforts to achieve this goal, a new design that involves cryptographic primitives and protocols and build systems based on them.
Catena: Efficient Non-equivocation via Bitcoin
Cadena is a system that uses Bitcoin’s security machinery to defend against online identity theft. An attacker who hacks a public-key encryption system might “certify” or cryptographically assert the validity of – a false encryption key, to trick users into revealing secret information. But it couldn’t also decertify the true key without setting off alarms, so there would be two keys in circulation bearing certification from the same authority. The new system defends against such “equivocation”.