DocumentCode
2028887
Title
Randomness in Virtual Machines
Author
Fernandes, Diogo A. B. ; Soares, Liliana F. B. ; Freire, M.M. ; Inacio, P.R.M.
Author_Institution
Dept. of Comput. Sci., Univ. of Beira Interior, Covilhä, Portugal
fYear
2013
fDate
9-12 Dec. 2013
Firstpage
282
Lastpage
286
Abstract
Virtualization technology provided cloud computing with the means to rapidly disseminate throughout the industry and achieve the utility computing long-envisioned era. Efforts on this research area have been focused on assuring isolation between co-resident virtual machines to avoid escaping the sandbox, but less attention has been given to the implications virtualization may pose to the efficiency and quality of random number generation on guests. On Linux distributions, the good provisioning of entropy gathered by the kernel is crucial for the functioning of its random number generator. However, entropy sources may be scarce on virtual machines due to the abstraction implied by the virtualization layer. As a consequence, both the generation speed and the quality of random numbers might drop when compared to hosts. This paper looks into this issue and analyzes the outputs of the /dev/random interface of the Linux kernel on virtual machines. With a well-know statistical library it is shown that the outputs are of high quality and are independently generated, even though they are produced on a slower basis.
Keywords
Linux; cloud computing; random number generation; software libraries; virtual machines; virtualisation; Linux distribution; cloud computing; co-resident virtual machines; entropy sources; random number generation; random number generator; statistical library; utility computing; virtualization layer; virtualization technology; Cloud computing; Entropy; Kernel; Linux; Security; Virtual machine monitors; Virtualization; Clouds; Randomness; Security; Virtualization;
fLanguage
English
Publisher
ieee
Conference_Titel
Utility and Cloud Computing (UCC), 2013 IEEE/ACM 6th International Conference on
Conference_Location
Dresden
Type
conf
DOI
10.1109/UCC.2013.57
Filename
6809414
Link To Document