Title :
Using Memory Map Timings to Discover Information Leakage to a Live VM from the Hypervisor
Author :
Marken, Brandon ; Hay, Brian
Author_Institution :
Dept. of Comput. Sci., Univ. of Alaska Fairbanks, Fairbanks, AK, USA
fDate :
June 27 2014-July 2 2014
Abstract :
In this paper we present a technique for detecting information leaked from the hypervisor to the guest OS via statistical analysis of the amount of time required to map and unmap a page in memory. Analysis of these timings allows a guest to determine the number of co-located VMs on the same physical hardware as well as determine whether or not the VM is being monitored by Virtual Machine Introspection.
Keywords :
operating systems (computers); paged storage; statistical analysis; virtual machines; colocated VM; guest OS; hypervisor; information leakage detection; memory map timing; page mapping; page unmapping; statistical analysis; virtual machine introspection; Hardware; Kernel; Monitoring; Timing; Virtual machine monitors; Virtual machining; Virtualization; Virtualization;
Conference_Titel :
Services (SERVICES), 2014 IEEE World Congress on
Conference_Location :
Anchorage, AK
Print_ISBN :
978-1-4799-5068-3
DOI :
10.1109/SERVICES.2014.18
