TaskShuffler++: Real-Time Schedule Randomization for Reducing Worst-Case Vulnerability to Timing Inference Attacks

TL;DR

This paper introduces TaskShuffler++, a real-time schedule randomization method that enhances security against timing inference attacks by increasing schedule unpredictability without compromising schedulability.

Contribution

It proposes a novel run-time schedule randomization algorithm and an information-theoretic measure to quantify worst-case schedule vulnerability.

Findings

Significantly reduces adversary success in predicting task execution times.

Maintains original schedulability while increasing schedule uncertainty.

Provides a new metric for assessing schedule security.

Abstract

This paper presents a schedule randomization algorithm that reduces the vulnerability of real-time systems to timing inference attacks which attempt to learn the timing of task execution. It utilizes run-time information readily available at each scheduling decision point to increase the level of uncertainty in task schedules, while preserving the original schedulability. The randomization algorithm significantly reduces an adversary's best chance to correctly predict what tasks would run at arbitrary times. This paper also proposes an information-theoretic measure that can quantify the worst-case vulnerability, from the defender's perspective, of an arbitrary real-time schedule.