Have you heard of or used VMware Flings before? The definition of fling on the website is: “A fling is a short-term thing, not a serious relationship but a fun one. Likewise, the tools that are offered here are intended to be played with and explored.” That’s just what I did with the newest version of vBenchmark, a fling that came out a couple weeks ago. vBenchmark lets you check for efficiency and baseline your VMware virtual environment in a really simple way.
To install vBenchmark, go to the labs.vmware.com/flings website and click on vBenchmark at the top of the page. Click “I agree…” and then select the .OVF file from the pull-down menu (assuming you’re going to put this virtual appliance on one of your ESXi hosts). Then click download. When this has finished downloading open your vSphere Client and go to File>>Deploy OVF Template. Browse to you downloaded .OVF file and select it. Go through the wizard picking the proper ESXi servers and datastores. There is an option to configure networking at the end of the wizard; however, I found that I had to reconfigure networking on the actual command line while it was running. Make sure you pick a network adapter that is compatible with Linux. Try pinging it to see if you have network connectivity.
If you do have network connectivity, you can open a browser and go to https://ipaddress.of.vBenchmark.server. You’ll need to enter your vCenter server IP address (or addresses) and credentials. Then click Initiate Query & Proceed to Dashboard. After a few seconds you will be taken to a dashboard that gives a brief overview of your initial environment. It shows various things like your physical and virtual infrastructure configuration, efficiency regarding CPU, memory, and administrator productivity, time it takes to provision or reconfigure VMs and quality of service.
One really neat thing you can do is compare this information with similar companies by clicking on the “Tell Me How” links next to each category. Essentially there is a repository of metrics classified by company size and industry. When you share your info (everything is scrubbed so as not to release any confidential data) the dashboard changes and you can compare your results with your peers (as you can see in Figure A). On the dashboard you may also click on “Metric Definitions” to get a clearer understanding of what all of these categories and metrics mean.
There are other tabs besides the Dashboard tab that allow you to go to the different categories individually. Under these tabs you can sort by vCenter instance or by vSphere edition. By default the values are averaged out, but you can also click on the “Show Minimum/Maximum” link to see more than just the averages.
Under the vSphere 5 RAM Pools tab it will show you a representation of how much vRAM you have and how much you’re entitled to (shown in Figure B). If you recall, VMware changed its licensing and each vSphere edition is entitled to a certain amount of vRAM. If you don’t remember what your edition includes, you can click on the “vSphere 5 vRAM Entitlements per Physical CPU” link to see the official vRAM entitlements for each edition.
I mentioned baselining in the first paragraph. Although this is kind of a quick and dirty way of doing it, there is an Options link in the upper right corner of the page where you can save or export your current data for review at a later date. This is kind of a nice feature to compare your current metrics to what you’ve come up with in the past. This is a really neat tool. The portal has a really clean and modern look to it. It’s easy to set up and comprehend and who doesn’t love to compare their environments to others. Check it out if you get a chance!