This is not the first time I write about Scale Computing but I met them again a couple of weeks ago (at #SFD5) and I think there are some updates about their technology and products that are worthwhile mentioning.

Scale computing has been around since 2009. At the beginning it was a scale-out storage particularly suited for SMB. Later, computing functionalities were added to it and now, with their 850+ customers and 1000+ deployments, they can be considered one of the most successful players in the hyper converged market so far.

But they are different

First of all, when compared to other names of hyper convergence, they target a different market. Scale computing solutions are good for customers who run in the range starting from a few VMs to less than 4 to 5 hundred… in other words, they target SMB!

scale_logoThe second important thing to know about Scale Computing is that it runs on the KVM hypervisor. Which is brilliant indeed.
In fact, especially for the SMB, where VMware’s most intriguing features are useless, running one hypervisor or another is the same. This kind of end user wants a simple, easy to use and stable virtualization platform with a set of features that simplifies life… once achieved, whether the job is carried out by ESXi or KVM is of no real importance.
Scale computing developed the whole the HC3 platform with their users in mind: easy to use web UI, templates and good support.

Third: probably the most important news comes from the storage layer. They have recently replaced GPFS with their SCRIBE.
This has two important consequences. They now have complete control of the entire software stack and there are no more licenses to pay to a third party. Yet another advantage for the customer who will be paying less for a better product. SCRIBE resembles Ceph under many concepts and looks quite impressive, at least on paper. (the video which can be found below, gives a clear idea about its design and implementation)

The problem of being different

Apart from some improvements that the product itself needs (like, for example, a native replica/DR option… on which Scale is working on for a future release, indeed), being different has its drawbacks too.
The first is the ecosystem of partners. I know Scale is building it, but you can count a very few partners when you look for complementary solutions (like for example a backup software).
I’m not saying that it is a problem for the majority of the end users, but I’m sure this is an aspect to consider when you look at these kinds of solutions.

Why it is important

Scale computing is not an alternative to other hyper converged products like Nutanix or Simplivity, it simply plays on a different field. This is a cost effective and well integrated solution that has found its place right where others have nothing to say or don’t want to be.

My personal POV is that the most important feature of Scale is not found in the technical components of its architecture (which are good, BTW) but in the fact that they own the whole stack and they can directly support it.
In other words: it just works, but if something goes wrong Scale’s support team know what has to be fixed! The kind of stuff that makes for a better user experience… and user experience is what SMB organizations care most about!

here a video with a demo of the product:

Disclaimer: I was invited to this meeting by Gestalt IT and they paid for travel and accommodation, I have not been compensated for my time and am not obliged to blog. Furthermore, the content is not reviewed, approved or published by any other person than the Juku team.