Paper 2014/713

Co-Location-Resistant Clouds

Yossi Azar, Seny Kamara, Ishai Menache, Mariana Raykova, and Bruce Shepherd

Abstract

We consider the problem of designing multi-tenant public infrastructure clouds resistant to cross-VM attacks without relying on single-tenancy or on assumptions about the cloud's servers. In a cross-VM attack (which have been demonstrated recently in Amazon EC2) an adversary launches malicious virtual machines (VM) that perform side-channel attacks against co-located VMs in order to recover their contents. We propose a formal model in which to design and analyze \emph{secure} VM placement algorithms, which are online vector bin packing algorithms that simultaneously satisfy certain optimization constraints and notions of security. We introduce and formalize several notions of security, establishing formal connections between them. We also introduce a new notion of efficiency for online bin packing algorithms that better captures their cost in the setting of cloud computing. Finally, we propose a secure placement algorithm that achieves our strong notions of security when used with a new cryptographic mechanism we refer to as a shared deployment scheme.

Metadata
Available format(s)
PDF
Publication info
Published elsewhere. Major revision. ACM Cloud Computing Security Workshop (CCSW) 2014
Keywords
cloud computingcross-VM attacksco-location attacksisolationco-location resistancebin packingsecure multi-worker delegation
Contact author(s)
senyk @ microsoft com
History
2014-09-12: revised
2014-09-12: received
See all versions
Short URL
https://ia.cr/2014/713
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2014/713,
      author = {Yossi Azar and Seny Kamara and Ishai Menache and Mariana Raykova and Bruce Shepherd},
      title = {Co-Location-Resistant Clouds},
      howpublished = {Cryptology {ePrint} Archive, Paper 2014/713},
      year = {2014},
      url = {https://eprint.iacr.org/2014/713}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.