Curiosity driven methods improve detection of hardware interference in multicore systems
Application of curiosity driven exploration methods for hardware interference identification
Artificial Intelligence
Summary
When computer chips have multiple cores that share parts, they can slow each other down in unpredictable ways. This makes it hard to guarantee that certain programs finish on time, which is especially important in planes and other safety devices. The authors treat this problem like exploring all possible weird slowdowns by using smart computer algorithms that are curious to find new behaviors. Their tests on a simulator show that these curiosity-driven methods find more types of interference more evenly than random tests can in the same time.
What this means in practice
- •For embedded systems engineers: Identify and characterize hardware interference effects in multi-core embedded platforms to ensure timing safety in avionics and related fields.
- •For ai developers in testing tools: Develop smarter test programs using curiosity-driven algorithms to more efficiently cover complex system behavior spaces when debugging hardware interactions.
Authors
Ludovic Matar, Clement Moulin-Frier, Pierre-Yves Oudeyer
Abstract
The transition from single-core to multi-core architectures in safety-critical embedded systems introduces significant challenges due to inter-core interference caused by contention for shared hardware resources. Such interference affects execution times and complicates the verification of strict temporal requirements, particularly in domains such as avionics where standards require comprehensive identification of interference sources. Existing interference analysis approaches, whether manual or model-based, struggle to capture the full range of behaviors arising from the complex interactions among micro-architectural components. In this paper, we frame multi-core interference analysis as the exploration of a complex system behavior space. We propose the use of curiosity-driven exploration algorithms from artificial intelligence to systematically and efficiently cover the space of possible interference behaviors. Using a simulator-based environment, we show that the proposed approach achieves broader and more uniform behavioral coverage within a limited experimental budget compared to traditional pseudo-random program generation methods.