Cloud computing performance benchmarking and virtual machine launch time

Authors:
Dylan Steinmetz;Brian W. Perrault;Ross Nordeen;Jacob Wilson;Xinli Wang
Affiliations:
Michigan Technological University, Houghton, MI, USA;Michigan Technological University, Houghton, MI, USA;Michigan Technological University, Houghton, MI, USA;Michigan Technological University, Houghton, MI, USA;Michigan Technological University, Houghton, MI, USA
Venue:
Proceedings of the 13th annual conference on Information technology education
Year:
2012

Citing 1
Cited 0

The Eucalyptus Open-Source Cloud-Computing System

CCGRID '09 Proceedings of the 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid

Quantified Score

Hi-index	0.00

Visualization

Abstract

This research is to study the performance of cloud computing platforms in the perspective of information technology (IT) management. Two separate test clouds of Eucalyptus and OpenStack were established on identical hardware. The BYTE UNIX benchmark suite was employed to conduct various performance tests on both clouds. While the OpenStack cloud out performed Eucalyptus considerably in the larger size copy test, the Eucalyptus cloud performed better than the OpenStack in the tests of serial excel and serial shell scripts. Scripts were written to compute the amount of time that was needed for the clouds to launch a virtual machine (VM) in two different ways. On average, it took a shorter time to launch a VM instance in both clouds when several VMs were created simultaneously than that when they were created one by one, the results showed a variation with different numbers of VMs that were launched concurrently.