Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Vulnerability Assessment of the Rowhammer Attack Using Machine Learning and the gem5 Simulator -Work in Progress

Loïc France 1 Maria Mushtaq 2 Florent Bruguier 1 David Novo 3, 1 Pascal Benoit 1
1 ADAC - ADAptive Computing
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
2 Lab-STICC_UBS_CACS_MOCS
Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance
Abstract : Modern computer memories have been shown to have reliability issues. The main memory is the target of a security attack called Rowhammer, which causes bit flips in adjacent victim cells of aggressor rows. Multiple mitigation techniques have been proposed to counter this issue, but they all come at a non-negligible cost of performance and/or silicon surface. Some techniques rely on a detection mechanism using row access counters to trigger automatic defenses. In this paper, we propose a tool to build a system-specific detection mechanism using gem5 to simulate the system and Machine Learning to detect the attack by analyzing hardware event traces. The detection mechanism built with our tool shows high accuracy (over 99.8%) and low latency (maximum 264µs to classify when running offline in software) to detect an attack before completion.
Liste complète des métadonnées

https://hal.umontpellier.fr/hal-03196090
Contributeur : Loïc France <>
Soumis le : lundi 12 avril 2021 - 14:34:50
Dernière modification le : mercredi 21 avril 2021 - 08:50:02
Archivage à long terme le : : mardi 13 juillet 2021 - 18:50:57

Fichier

ACM_RH_mitigation(2).pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-03196090, version 1

Citation

Loïc France, Maria Mushtaq, Florent Bruguier, David Novo, Pascal Benoit. Vulnerability Assessment of the Rowhammer Attack Using Machine Learning and the gem5 Simulator -Work in Progress. ACM Workshop on Secure and Trustworthy Cyber-Physical Systems (SaT-CPS 2021), Apr 2021, Virtually, United States. ⟨hal-03196090⟩

Partager

Métriques

Consultations de la notice

106

Téléchargements de fichiers

200