Building a private cloud infrastructure tends to be a cumbersome process: even if you do it right, you oft have to deal with four to six different components: orchestration system, hypervisors, servers, storage arrays, networking infrastructure, and network services appliances.

As I explained in the Designing Private Cloud Infrastructure webinar and several blog posts, you can replace most networking hardware with VM-based solutions, and use distributed file systems to replace storage arrays (you’ll find a Nutanix-based case study in my webinar and Data Center Design Case Studies book).

After launching VSAN a year ago (and having some interesting field experiences), VMware decided to enter fray for real and announced EVO:RAIL – a tightly integrated software and hardware solution targeting mid-sized cloud deployments.

What is EVO:RAIL?

In a nutshell: a hardware appliance (= blade enclosure) with four blades (nodes) running vSphere and VSAN. Each blade has up to 2 CPUs (12 cores), 192 GB of memory, 3.6TB of disk, two 10GE and one 1GE port.

The hardware details are a bit sketchy at the moment: a parameter that is sometimes quoted as “up to” appears as “at least” in other sources of information. We’ll have to wait till the actual hardware appliances are launched to get these details right.

How is it different from Nutanix?

Good question ;) The fundamental difference is tight integration with vSphere and vCenter. The EVO:RAIL user interface (a layer hiding the complexities of vCenter from the innocent eyes of an SMB administrator) makes installation, cluster expansion and basic administration extremely easy… and you can always fall back to vCenter to do the hard-core configuration. For more eye candy, watch the video on VMware’s web site.

Why would I still buy Nutanix?

VMware is starting with a pretty conservative hardware configuration. A high-end Nutanix node in the same form factor has 20 cores and 512GB of memory, giving you at least twice as many VMs in the same physical space.

Nutanix solution seems to be more scalable at this moment. The initial release of EVO:RAIL supports up to four appliances (16 nodes), for a maximum of 16 vSphere hosts, ~400 server VMs or ~1000 VDI VMs (numbers straight off VMware data sheet). The Nutanix Starter edition is a rough equivalent with 12 nodes in a cluster (but with more capable nodes), the Pro- and Ultimate editions go way beyond that number.

Why would I buy VMware EVO:RAIL?

In a word: single point of blame ;) It goes against everything HP, Gartner, SDN evangelists, and whitebox switching aficionados tell us, but sometimes it makes more sense to pay more and have peace of mind (not to mention uptime) than to troubleshoot the hidden intricacies of home-brewed concoctions.

What can possible go wrong?

VMware would upset too many partners (and its majority owner) if it would start selling VSAN-based converged hardware appliances. EVO:RAIL is thus a solution that will be sold by VMware hardware partners (VMware promises a single SKU and with a single support contract). Let’s see how well that works out (and looking at the list of partners one has to wonder when EMC became a server vendor ;).

More information

To get more EVO:RAIL technical details, read in-depth blog posts by Chris Wahl and Duncan Epping (looks like EVO:RAIL was his baby – CONGRATULATIONS!!!).

For a bigger picture watch my Designing Private Cloud Infrastructure webinar and other cloud-related webinars, read my Data Center Design Case Studies book, or attend my Interop New York sessions. We can also review your cloud infrastructure plans and designs, or review potential technology options in a short online consulting session.