Let's Stop Trusting Software With Our Sensitive Data
Tuesday, October 15, 2013
Nary a day goes by without hearing about break-ins into software systems, with personal or confidential data being compromised. Yet, as time goes on, we are trusting the cloud more and more to perform sensitive operations for us. Demanding more trust in software systems is a recipe for disaster.
Hardware to the rescue! Suppose we only trust hardware manufacturers and cryptographers, and not system software developers, application programmers, or other software vendors. It will be the hardware manufacturer's job to produce a piece of hardware that provides some security properties. The additional physical security that comes with hardware is a bonus; however, there is still a leap of faith! We must trust that the hardware's security guarantees really do take software out of the loop.
This poses a challenging problem. Software that operates on our data is assumed to be curious or malicious. To make matters worse, the cloud service provider can also be malicious and can run whatever program it wants on our data. How can we ensure privacy of data despite the practically infinite number of malicious programs out there?
We describe the architecture of the Ascend (Architecture for Secure Computation on Encrypted Data) processor that achieves these goals while running batch or stream computations that operate on encrypted client data and return encrypted results to the client; the only entity that the client has to trust is the processor itself. Ascend ensures privacy even though the adversary has control over what software runs on Ascend, and can monitor all external pin traffic.
Srini Devadas is the Edwin Sibley Webster Professor of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT), where he has been on the faculty since 1988. He served as the Associate Head of EECS with responsibility for Computer Science from 2005-2011. Devadas has worked in the areas of Computer-Aided Design, testing, formal verification, compilers for embedded processors, computer architecture, computer security, and computational biology and has co-authored numerous papers in these areas and received several best paper awards. Devadas was elected a Fellow of the IEEE in 1999.
Host: Hank Hoffmann
Argonne National Laboratory
The Chicago Center for the Theory of Computing and Allied Areas
Please check out our Distinguished Lecture Series