Scale Computing HC3 Cloud Unity – Whats that about then?

Background.

I recently returned from Tech Field Day 15 where Scale Computing was giving the delegates an overview of their very recently announced HC3 Cloud Unity with Google Cloud Platform. I will be honest, up until setting foot in the fabulous Scale Computing office in San Francisco, all I knew about the company was the name and that they play in Hyper-Converged Infrastructure space, having coined the term themselves in 2012. What I quickly learned was that this solution is based on KVM hypervisor and is driven and managed from a simple to use web-based user interface. The simplicity of it all is what grabbed my attention.

Whats the premise?

In a nutshell, run virtual machines seamlessly both on-premises and in Google Cloud Platform without having to change IP addresses or any jazz like that. Pretty cool don’t you think?

How does it work?

The cross-platform part of this ties in nicely with Googles own announcement to allow nested virtualisation on their Google Cloud Platform. Scale Computing and Google worked together for 2 years prior to bringing this solution to the market. Check out the link from Google and you will see Scale Computing mentioned.

Nested virtualisation may sound like a crazy thing to do, but look at it this way, it allows Scale to operate their HC3 virtualisation platform as an Infrastructure as a Service (IaaS) model in Google Compute Engine, to then allow you to run workloads you currently run on premises in the “cloud” without the need to perform any kind of virtual machine conversion. Put simply, you can move virtual machines from on-premises location to the cloud without any hassle. Great for a disaster recovery scenario. Your on-premises infrastructure blows up, no problem, just power it back on again in the cloud!

What makes this solution even simpler is that Scale Computing utilises VXLAN encapsulation to create a stretched layer 2 network between the two locations. What this means for none networking types like myself is you do not need to change the IP address of any virtual machines if you move them between locations. It may make failover testing tricky though if the failover location network cannot be isolated from the production network, that is something that requires further investigation. Disaster recovery testing is mentioned as a line item as part of the HC3 Cloud Unity DRaaS in the FAQ though, so I guess it is possible.

The VXLAN is facilitated by a small virtual machine which can be spun up anywhere to allow access to the GCP instance from the network that the virtual machine is running on.

So what data is transferred to the cloud? If you define a snapshot schedule on your on-premises workloads, those snapshots are then transported to the Google Cloud Platform after the snapshot is taken. I believe it is a forever incremental snapshot chain after the initial sync.

There is the option for passive failover for testing or you can log a support call to initiate full disaster failover recovery in active mode. The resources available in GCP will be increased in the event of a full failover.

Thoughts.

Now I said this seems very simplistic, which it is, but to a point that this needs some additional functionality to allow the solution to scale in larger environments. There is no native sandboxing of the GCP instance. There is no orchestration at this point for failover to GCP although an API is in the works to allow for this sort of thing.

Coming from a VMware background though, it is refreshing to see a solution that should just work out the box. I am confident there any many environments that don’t need all the bells and whistles that vSphere and a VMware cloud service provider of some description can offer.

Further learning.

There are a bunch of white papers and technical resources on Scale Computing’s HC3 platform available on their website if you would like to take a deep dive into the technology. There is also a dedicated page for Scale Computing HC3 Cloud Unity here.

FAQ Document here with indicative costing.

Scale Computing is the conduit for all costs for the service, including the costs incurred with any IaaS running in Google Cloud Platform.

Check out the Cloud Unity overview video below recorded at Tech Field Day 15.

The cover your ass bit.

As this blog post is on the back of a Tech Field Day event, it would be fair to state that my travel, accommodation, food etc was covered for the duration of the event. The event is paid for by the vendors sponsoring the event. Opinions, however, are my own and I have not made any financial gain by writing this blog post.

Ian