DATE:Thursday, October 27, 2022
TIME:12:00 - 1:00 pm, to be presented via Zoom
PLACE: A zoom link will be sent to the SDI mailing list a day before the talk.
SPEAKER: Sam Kumar, UC Berkeley
TITLE: MAGE: Nearly Zero-Cost Virtual Memory for Secure Computation
ABSTRACT:
Secure Computation (SC) is a family of cryptographic primitives for computing on encrypted data in single-party and multi-party settings. SC is being increasingly adopted by industry for a variety of applications. A significant obstacle to using SC for practical applications is the memory overhead of the underlying cryptography. We develop MAGE, an execution engine for SC that efficiently runs SC computations that do not fit in memory. We observe that, due to their intended security guarantees, SC schemes are inherently oblivious --- their memory access patterns are independent of the input data. Using this property, MAGE calculates the memory access pattern ahead of time and uses it to produce a memory management plan. This formulation of memory management, which we call memory programming, is a generalization of paging that allows MAGE to provide a highly efficient virtual memory abstraction for SC. MAGE outperforms the OS virtual memory system by up to an order of magnitude, and in many cases, runs SC computations that do not fit in memory at nearly the same speed as if the underlying machines had unbounded physical memory to fit the entire computation.
A paper describing this research appeared at OSDI 2021, where it received a Jay Lepreau Best Paper Award.
BIO:
Sam is a final-year PhD student at UC Berkeley, advised by David Culler and Raluca Ada Popa. He is interested in system security and networked systems. Sam's research focuses on rethinking system design for expressive cryptography. More information is available at his website.
SDI SEMINAR QUESTIONS?
Karen Lindenfelser, 86716, or visit www.pdl.cmu.edu/SDI/