Issue No. 03 - May-June (2014 vol. 18)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MIC.2013.104
Kejiang Ye , Zhejiang University, China
Zhaohui Wu , Zhejiang University, China
Bing Bing Zhou , University of Sydney, Australia
Xiaohong Jiang , Zhejiang University, China
Chen Wang , Commonwealth Scientific and Industrial Research Organisation, Australia
Albert Y. Zomaya , University of Sydney, Australia
Modern cloud computing environments often use virtualization technology to implement resource sharing. However, virtualization incurs performance overheads due to the additional hypervisor layer placed between virtual machines (VMs) and hardware resources. Understanding VM performance is crucial for running applications efficiently and for further improving VM techniques. Unfortunately, existing benchmarking methods are insufficient for comprehensively measuring virtualization performance. Here, the authors propose a solution that fully evaluates VM system performance. In addition to discussing their three-layer benchmarking methodology, they present a new benchmark suite -- Virt-B -- that measures various virtualization scenarios. They also introduce a performance testing toolkit to automate the benchmarking process and describe three case studies.
Benchmark testing, Virtual machining, Virtualization, Virtual machine monitors, Cloud computing, Resource management
K. Ye, Z. Wu, B. B. Zhou, X. Jiang, C. Wang and A. Y. Zomaya, "Virt-B: Towards Performance Benchmarking of Virtual Machine Systems," in IEEE Internet Computing, vol. 18, no. 3, pp. 64-72, 2014.