Nearly two years ago I was given the opportunity to take out the Condusiv Technologies V-locity VM version 4 product for a spin in my home lab, the results were pretty positive and certainly something I would suggest to anyone running a Windows estate to investigate.
Move forward a couple of years and I was contacted by their technical director after having filled out an online questionnaire, he remembered my blog articles and wondered if I wanted to take the new version of V-locity 5 out for a spin to see how it compared, V-locity 5 now combines both the V-locity VM and V-locity Server products under one product banner.
V-locity is an I/O reduction software solution that increases application performance in virtual, physical and cloud environments by 50% or more without the cost of additional hardware
In this version of V-locity you’re provided with the VMC (V-locity Management Console) which provides a method of deploying the agent to any domain joined Windows machine (Windows 7, 8, 8.1, Windows Server 2003, 2008 (and R2), 2012 (and R2), if you don’t like that you can also create a package for distribution via a 3rd party Systems Management tool (SCCM 2012R2 for example).
I am interested to see how much improvement there is between the two versions so I will be running a side by side comparison.
During the next couple of weeks I will be completing blog posts about installing the VMC, deploying agents and comparing the performance differences of V-locity 5, V-locity VM and a base OS using both Server and Desktop Operating Systems.
Of course over the years technology changes so this time around my testing will be carried out on the following hardware.
ESXi HOST: Shuttle XPC SH67H3, 32GB RAM; 1x i7 3770 @ 3.4ghz with ESXi 6.0 (6.0.0 build 2715440) in a 3 node cluster with DRS rules present.
Synology: DS1513+ (5 x Samsung EVO 500GB SSD) – Configured for iSCSI Multipathing with a total of 9 paths per server using the Round Robin PSP.
NetGear: TS724T 24 x 1 GB Ethernet switch
NetGear: TS724T 24 x 1 GB Ethernet switch (dedicated storage switch for the Synology unit).
IO Test Machines
OS TYPE: 3 x Windows 7 x64 (fully patched) on ESXi 6.0 using a 40gb thick eager zero provisioned disk (1 x Standard VM, 1 x V-locity VM and 1 x V-locity 5)
OS TYPE: 3 x Windows 2012R2 (fully patched) on ESXi 6.0 using a 60gb thick eager zero provisioned disk (1 x Standard VM, 1 x V-locity VM and 1 x V-locity 5)
CPU Count \ Ram: 2 vCPU, 4GB ram (It’s been suggested to run the VM’s with 6GB of ram as that offers a better available cache, to see what the performance difference will be I will run the tests twice, the first time with the usual 4GB of ram, the second set of runs will be allocated 6GB each, my preference for Windows boxes tends to be 4GB but I want to see if the extra 2GB pays off).
These were templated VMs with full updates applied. Two guests are running V-locity VM and Iometer via a PowerShell script, two more were running V-locity 5 and Iometer via the PowerShell script, the last two guests were only running Iometer via the PowerShell script. In addition the rest of my estate was running on the hosts and storage to provide an environment under normal load.
The power plan was set to High performance and the screen saver turned off for all VMs.
IO Testing Software
All testing will be carried out using IO Meter software and test scripts provided by Condusiv. Testing will be carried out over 4 days to allow the V-locity Benefits Analyser and Benchmark Data Collection (V-locity 5) to run to completion successfully (3 phases of testing – Analysis, Improvement and Post).
The IOMeter test script can be downloaded from here and the IOMeter.ps1 PowerShell file can be downloaded from here. No deviation to the testing script occurred.