CAREER: Integrating Microarchitecture Simulation and Side-Channel Leakage Modeling for Safer Software

University Of California-Los Angeles, Los Angeles CA

Investigators

Abstract

Computers, ranging from embedded systems to servers, are becoming increasingly pervasive and critical in our lives. Whilst these systems carry out their main function, which is computation, they inadvertently generate traces that could potentially disclose sensitive information through alternative physical or digital pathways known as “side-channels.” As we become increasingly dependent upon computing systems, it is more critical than ever to know how side-channel signals can be created, how side-channel leakage can be modeled, and how future software systems should be designed to be robust against side-channel attacks. This proposal explores new methods and tools for quantifying and modeling side-channels for a wide range of processors. Successful completion of this project brings new methods, findings, and open-source tools for analyzing and quantifying side-channels. Various designers and industries can gain advantages from these techniques and tools, including hardware designers, computer architects, compiler designers, and software developers. Our findings enable the construction of secure software systems while reducing cost and time-to-market. As an integral part of this research program, we also propose an educational agenda involving K-12, undergraduate, graduate, and broader security community education. The overall goal of this research proposal is to build a comprehensive tool for side-channel leakage analysis. To achieve this, we first need to improve our modeling capabilities, particularly for physical side-channels, and then focus on integrating these models into established simulation and analysis tools. Based on these needs, our proposal is structured into two research thrusts and an evaluation plan. Each thrust is further broken down into three main research objectives. The first thrust is focused on building an accurate tool to estimate power and electromagnetic (EM) side-channels at the software level. Using three research objectives, we plan to build a model that can accurately model various processor architectures as well as other important components (e.g., accelerators) on a system-on-chip (SoC). The second thrust is focused on modeling microarchitectural side-channels. Specifically, the main goal is to build a new tool that can combine established binary analysis tools with known side-channel leakage models. The ideas proposed will be evaluated using a detailed evaluation plan. We will apply our methods to a wide variety of systems and setups with different hardware and software designs. We will consider various scenarios including different microarchitecture designs (in-order, out-of-order, multicore) and different side-channels (physical electromagnetic signals, digital cache side-channels, transient/speculative side-channels). This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.