How we tested the virtualization products

We used the same host platform, an HP DL580 G5 (four-socket, 16-core Intel Xeon CPUs) server – for the qualitative portion of this test as we did in the quantitative portion of our test published earlier this month.

We researched hypervisor platform compatibility lists to make our compatibility assessment.

We tested the importation/migration and setup of virtual machines using both native hypervisor applications (Microsoft's Hyper-V 1.0 and ESX 3.5) as well as Microsoft's beta/RC of Systems Center-Virtual Machine Manager and VMware's VirtualCenter 3.5. We used an IBM x3550 (two-socket, eight-core Intel Xeon V-enabled CPUs) and Dell 1950s (two-socket, eight-core Intel V-enabled CPUs) using local SAS storage as base platforms to test both single and SMP kernel migrations from the HP DL580 server as well as for cloning of Windows 2003 and 2008 Server Editions as well as Novell SUSE Linux Enterprise Server V 10.2 in monolithic and SMP kernel configurations.

We then monitored virtual machines through differing stages of guest behavior from startup to shutdown while watching activity. We then tested and noted access methods used for Microsoft's SC-VMM and VMware's Virtual Center to note the relative accessibility of the hypervisors and how firewalling and API/logon access may provide undesired security holes that could be cracked.

Return to test: The virtual winner: VMware’s ESX KOs a roughly built Hyper-V package.