As I saw the news from IDC on how they are expanding their definition of hyperconverged systems to include those that are disaggregated in nature such as our NetApp HCI, I immediately started thinking of a blog and what the title would be and the one above got stuck in my head. I apologise if you know where this came from and like me, you’ll now also have one of the worst songs ever written stuck in your head too. If you don’t then let me apologise and present to you The Firm – Star Trekkin’
So what just happened and why am I writing this?
At the end of October 2017 we launched NetApp HCI, what was different though was that we made a very conscious design choice in that we wanted to give customers the ability to scale compute and storage independently but continue to provide the simplicity that had been at the core of the HCI proposition.
And so began the 20 months of, is it really HCI or is it CI? As bloggers, analysts and competitors desperately tried to fit it into one category or another, or dismiss it from one category or another, which ever made for the best article or best outcome in a competitive situation.
NetApp were pretty clear from the outset in that we believed what we were creating was an evolution of HCI. We designed it to have the simplicity and other benefits that the generation 1 HCI solutions had delivered but added on this ability to be much more flexible in how compute and storage could be balanced and scaled, which tends to be much better suited to enterprise applications where the ratio of scaling is not consistently linear.
Yes there were occasions where we were simply dismissed as we didn’t meet the analysts definitions of what HCI was and wasn’t, but equally we had a significant number of customers that recognised the value of what we had built and saw how it could address use cases that generation 1 HCI systems couldn’t.
Today IDC have announced that they are expanding their definition of HCI. In addition to the more traditional definition they are now also including what they specify as ‘Disaggregated Hyperconverged Systems’ and they specifically call out the NetApp implementation as an example. They say that this is in response to the maturing market which is seeing a natural evolution toward new architecture types.
I believe this is a good first step in the right direction, but I do realise that this is just one of the analysts, albeit one of significance. There’s still a way to go to get others with their Wizardry Cubes to also acknowledge this evolution but I’m a firm believer that customer behaviour changes analyst views and as we continue to see the uptake of ‘Disaggregated Hyperconverged Systems’ then it’s just a matter of time before it gets widespread acknowledgement.
I am once again very proud of my colleagues that designed and developed NetApp HCI and asked people to challenge the current norms.
More info on NetApp HCI can be found here
And the IDC update here
HCI value is DataCenter in a box, based on simplicity, automation and scalability. The CPU/storage in same box is a technical answer and IDC took it as definition. We provided new approach and they call it disaggregated. Fine, this is recognition from market we are right.
So I guess our opinions have changed a little in the last 10 months. But only on a technology level. I bet however we still agree on the customer requirement. It really shouldn’t matter how the hardware is designed. Hardware just needs to be reliable and resilient. The storage type shouldn’t matter, but it needs to protect data from loss and provide typical functionality such as replication, snapshots clones etc and the hypervisor doesn’t matter as long as it provides the ability to spin up VMs and provide resilience etc. You don’t ask AWS 20 questions about hardware and software when you provision a VM instance and that’s what customers now want. Invisible infrastructure….. and that means it’s software driven. That means a cloud…. be it public or behind your own firewall or both delivered as one single cloud…. yes “hybrid” What customers are striving for is a system that you simply request a VM and the software decides on where to place that VM based on criteria such as Performance, geo location, security, technical ability and cost. The software makes the decisions and places the VM appropriately and in say 10 days time checks all those variables again and moves the VM if required. The automation of all that is where the real value is. Let’s not scrap about your HCI is not HCI and ours is….. if we are going to have a scrap lets do it over something that can deliver better outcomes. The software stack…. the use-ability and the simplicity. (Typed on my phone)
Didn’t really see any scrap coming or happening here Mick, long comment from you and not really sure what point you’re making. Unless it’s that technology is about solving customer problems in which case I agree that both our companies do, just in different ways.
Thanks for the comment though
Yea that’s it really…. when I say scrap I’d mean a sensible discussion. hope all is well with you 🙂