Abstract
Virtualization allows creating isolated computing environments, often called virtual machines (VM), which execute on a shared underlying physical hardware infrastructure. From a application designer’s perspective, it is critical to understand performance characteristics of such virtualized platforms. This is a challenging task mainly due to: a) There are multiple types of virtualization platforms such as Virtual Machine Monitor (VMM) based and operating system (OS) based. b) Most existing studies are focussed on evaluating narrow low-level subsystems of such platforms.
Three main forms of virtualization platforms – bare metal VMM, hosted VMM and OS based – are examined to understand their performance characteristics, such as CPU and memory usage, cross VM interference, from different dimensions. These characteristics are examined at macro level from applications perspective by subjecting these platforms to different types of load mix. One of our findings shows that the impact of VM co-location on different performance indicators is strongly dependent on platform and workload types.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Barham, P., et al.: Xen and the art of virtualization. SIGOPS Oper. Syst. Rev. 37, 164–177 (2003)
Rosenblum, M.: The reincarnation of virtual machines. Queue 2, 34–40 (2004)
Lezcano, D.: lxc linux containers, http://lxc.sourceforge.net/ (retrieved March 2012)
Ahmad, I., et al.: An analysis of disk performance in vmware esx server virtual machines. In: 2003 IEEE International Workshop on Workload Characterization, WWC-6 (2003)
Menon, A., et al.: Diagnosing performance overheads in the xen virtual machine environment. In: Proceedings of the 1st ACM/USENIX International Conference on Virtual Execution Environments (2005)
Padala, P., et al.: Performance evaluation of virtualization technologies for server consolidation. HP Laboratories Technical Report (2007)
Kolyshkin, K.: Virtualization in linux (openvz white paper) (2006), http://download.openvz.org/doc/openvz-intro.pdf
RUBiS Team, Rubis: Rice university bidding system. RUBiS Project, http://rubis.ow2.org/ (retrieved March 2012)
Watson, J.: Virtualbox: bits and bytes masquerading as machines. Linux Journal 2008 (February 2008), http://dl.acm.org/citation.cfm?id=1344209.1344210
Tomcat Project, Apache tomcat. The Apache Software Foundation, http://tomcat.apache.org/ (retrieved March 2012)
Hillyer, M.: Sakila sample database version 0.8. MySQL AB, http://dev.mysql.com/doc/sakila/en/sakila.html (retrieved March 2012)
JMeter Project, Apache jmeter. The Apache Software Foundation, http://jmeter.apache.org/ (retrieved March 2012)
GNU Octave, Statistics analysis, GNU, http://www.gnu.org/software/octave/doc/interpreter/Statistics.html (retrieved March 2012)
kernel.org, Linux profiling with performance counters. The Linux Kernel Archives, https://perf.wiki.kernel.org/ (retrieved March 2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sodhi, B., Prabhakar, T.V. (2012). Performance Characteristics of Virtualized Platforms from Applications Perspective. In: Hameurlain, A., Hussain, F.K., Morvan, F., Tjoa, A.M. (eds) Data Management in Cloud, Grid and P2P Systems. Globe 2012. Lecture Notes in Computer Science, vol 7450. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32344-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-32344-7_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32343-0
Online ISBN: 978-3-642-32344-7
eBook Packages: Computer ScienceComputer Science (R0)