LightRoAD

Lightweight Rowhammer Attack Detector

Conference Paper (2021)
Author(s)

M. Taouil (TU Delft - Computer Engineering)

Cezar Rodolfo Wedig Reinbrecht (TU Delft - Computer Engineering)

Said Hamdioui (TU Delft - Quantum & Computer Engineering)

Johanna Sepulveda (AirBus Defence and Space GmbH)

Research Group
Computer Engineering
Copyright
© 2021 M. Taouil, Cezar Reinbrecht, S. Hamdioui, Johanna Sepulveda
DOI related publication
https://doi.org/10.1109/ISVLSI51109.2021.00072
More Info
expand_more
Publication Year
2021
Language
English
Copyright
© 2021 M. Taouil, Cezar Reinbrecht, S. Hamdioui, Johanna Sepulveda
Research Group
Computer Engineering
Bibliographical Note
Accepted author manuscript@en
Pages (from-to)
362-367
ISBN (print)
978-1-6654-3947-3
ISBN (electronic)
978-1-6654-3946-6
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Dynamic Random Access Memory (DRAM)-based systems are widely used in mobile and portable applications where low-cost and high-storage memory capability are required. However, such systems are prone to attacks. A latent threat to DRAM-based system security is the so-called Rowhammer attacks. By repeatedly accessing memory, an attacker is able to perform unauthorized data modifications into physically adjacent memory locations. As a consequence, powerful privilege-escalation attacks can be achieved. Although most of the known countermeasures are based on refresh strategies or intensive address monitoring, their efficient and low-cost realization is still a challenge. In this work, we present LightRoad, a lightweight and flexible hardware detector for Rowhammer attacks. Additionally, we propose two variants that further extend the LightRoad security, namely LightRoAD+Sec and LightRoAD+PARA. Our experiments show that LightRoad and its variants are very efficient and effective to detect attacks while having an affordable cost that varies according to the desired security level.

Files

ISVLSI21_HAMMER_1_.pdf
(pdf | 0.338 Mb)
License info not available