Optimized Memory Encryption for VMs Across Multiple Hosts

• Main Authors: Horio, S., Takahashi, K., Kourai, K., Rahim, L.A.
• Format: Article
• Published: Springer Science and Business Media Deutschland GmbH 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85113731740&doi=10.1007%2f978-3-030-84910-8_32&partnerID=40&md5=bd6babf964949ee53aae6edf7c4c2bfc; http://eprints.utp.edu.my/28844/

Recently, virtual machines (VMs) with a large amount of memory are widely used. It is often not easy to migrate such a large-memory VM because VM migration requires one large destination host. To address this issue, split migration divides the memory of a VM into small pieces and transfers them to multiple destination hosts. The migrated VM exchanges its memory data between the hosts using remote paging. To prevent information leakage from the memory data in an untrusted environment, memory encryption can be used. However, encryption overhead largely affects the performance of the hosts and the VM. This paper proposes SEmigrate for optimizing the memory encryption in split migration and remote paging. SEmigrate avoids decrypting memory data at most of the destination hosts to reduce the overhead and completely prevent information leakage. Also, it selectively encrypts only the memory data containing sensitive information by analyzing the memory of the guest operating system in a VM. SEmigrate could reduce the CPU utilization during encrypted split migration by 6�20 point and improve the performance of the migrated VM with encrypted remote paging to 1.9 �. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.