DocumentCode :
1711024
Title :
Toward Automatically Deducing Key Device States for the Live Migration of Virtual Machines
Author :
Guodong Zhu ; Kang Li ; Yibin Liao
Author_Institution :
Dept. of Comput. Sci., Univ. of Georgia, Athens, GA, USA
fYear :
2015
Firstpage :
1025
Lastpage :
1028
Abstract :
The ability of migrating running virtual machines (VMs) in cloud environment provides significant benefits in dynamic resource load balancing and higher fault tolerance. The migration of a virtual machine consists of both the application/OS memory state migration and virtual hardware device state migration. Failures in the either of these two may lead to unpredictable behavior of the running VM. Previous researches have focused on the consistence of the application/OS memory state. In this paper, we inspect the migration of device states, which are also essential for the success of VM live migration. The current practice of defining device states is done by the developers of each virtual device and thus is prone to errors. We present an approach that automatically deduces what states are critical for a virtual device. Having the precise states defined is critical for the success of VM live migration.
Keywords :
cloud computing; operating systems (computers); resource allocation; software fault tolerance; virtual machines; VM live migration; application-OS memory state migration; cloud environment; dynamic resource load balancing; fault tolerance; key device states; virtual hardware device state migration; virtual machine live migration; Cloud computing; Conferences; Feature extraction; Hardware; Virtual machining; Virtualization; Live Migration; Virtual Device; Virtualization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Cloud Computing (CLOUD), 2015 IEEE 8th International Conference on
Conference_Location :
New York City, NY
Print_ISBN :
978-1-4673-7286-2
Type :
conf
DOI :
10.1109/CLOUD.2015.143
Filename :
7214154
Link To Document :
بازگشت