Cross-layer instruction-aware timing error mitigation & evaluation for energy-efficient dependable architectures

Read full text Summarise with VIRAT
Authors: Ioannis Tsiokanos
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Increased variability renders nanometer circuits extremely prone to timing errors that threaten system functionality and reliability. To protect circuits from timing errors, designers adopt pessimistic timing margins, which lead to energy inefficiency. This dissertation focuses on addressing the challenges related to energy efficiency and timing errors in a collective fashion. The rate and impact of such errors depend on their manifestation across the layers of the application, microarchitecture and circuit. Accordingly, this thesis investigates cross-layer methods to mitigate, evaluate and model timing errors, exploiting the data-dependent timing behaviour of pipelined designs. In the first part of this thesis, techniques that minimise, detect, and prevent timing errors are proposed. At the circuit-layer, this thesis investigates the root causes of timing errors and proposes a framework that isolates the timing critical paths to a single pipeline stage. At microarchitecture-layer, a dynamic cycle adjustment technique is devised to prevent timing errors in case of excitation of a timing critical path. At application/software-layer, the concept of approximate computing is leveraged to minimise timing errors. In the second part, two accurate timing error modeling and evaluation frameworks are proposed; for the first time the instruction execution history (i.e., type and order of instructions within a pipeline at any instant) is considered. DEFCON, a fully automated framework which customises a genetic algorithm driven by accurate dynamic timing analysis to stochastically search for microarchitecture-aware instructions that trigger timing errors, is presented. ARETE is then derived, a novel framework that enables fully-accurate impact-evaluation of timing errors on applications by combining dynamic binary instrumentation with machine learning-guided dynamic timing analysis. Finally, the inherent complex dynamic timing behaviour of pipelined architectures is exploited and a low power security primitive for hardware-rooted device authentication is proposed. To achieve this, DTA-PUF, a novel lightweight physical unclonable function, is introduced.

Fields of Study & Subjects

Low-power high-performance VLSI design Radiation Effects in Electronics Physical Unclonable Functions (PUFs) and Hardware Security

Publication details

Journal / source
Research Portal (Queen's University Belfast)
Publication date
1 October 2026
Language
EN
Document type
Thesis