The model is inverting.
This is how Isaiah Weiner, Principal Technologist at AWS, summarized the state of IT infrastructure during our recent virtual panel. It’s a good way to put it. In the weeks since my last post, as I’ve continued to meet virtually with enterprise IT leaders and executives, a visible trend is emerging.
We are in the age of distributed disasters, and there is a growing awareness now that relying on technology that has physical dependencies, even recovery-themed technologies, can suddenly become a liability. This is the inverted model Isaiah refers to, and it signals an enormous change. We’ve depended on physical infrastructure for so long. Enormously successful businesses have been built on a foundation of spinning disks and expensive iron packed in offices and data centers.
It used to be that being able to put your arms around the equipment that ran the technology was reassuring. Today, any infrastructure we can put our arms around has become a liability.
Enterprise IT leaders are realizing that their organizations would be much better off if the functions long carried out by local hardware were instead managed at scale by cloud giants like Azure and Amazon. Enterprise IT is now looking to shift as much infrastructure as they can to the cloud.
Yet success is not a given. Enterprises recognize the value in shifting their infrastructure to the cloud, especially in these unusual times, but still no one is going to tolerate poor performance. Not large global companies. Certainly not end users who are accustomed to working at the speed of thought. So what will it take for this all-cloud infrastructure of the future to take hold?
Innovative, Proven Cloud Giants
First and foremost, this new model depends on established cloud operations like Amazon and Azure. If all infrastructure goes to the cloud, then enterprises need to engage with cloud partners that can duplicate infrastructure completely – everything from directory services to load balancers, networking, security, and more – and have a truly global footprint capable of delivering strong performance everywhere around the world.
Nasuni is going to be there for file services, and we are working behind the scenes on enhancements that will establish Nasuni as the platform of choice for this all-cloud infrastructure of the future. The virtual machines we run in the cloud need to equal or exceed the performance of local hardware, and we need to be able to deploy more powerful virtual machines that can handle more users and bigger loads.
One of our primary advantages, relative to this all-cloud future, is that in designing our platform as cloud-native, we are equipped to operate in these environments. We can deal with increased load in a way most storage systems cannot. For instance, Nasuni operates through a single global file system, and all of our appliances can see the same file system, so when load increases, we can adapt and use more compute resources quickly in order to mitigate any performance degradation.
This will not be a small shift for us. Our focus as an organization has been getting hundreds of virtual and physical on-prem appliances to work well together at a global scale. Now we’re looking at optimizing a dozen or two strategically placed, high-performance cloud appliances that support virtual infrastructures. Thankfully, because of our cloud-native roots, we’re ready for the shift. The legacy hardware vendors we compete with cannot say the same thing.
Truly Scalable Cloud Services
Any vendor can ship their boxes to Microsoft or AWS and call themselves cloud. But cloud is not a place. It is not a location. It’s an architecture that allows for automation and scale, a business model that lets you scale up and down like a utility. These features are far, far more important than whether you have a NetApp box in your own data center or one belonging to Azure or AWS. If you take some NetApp box and Microsoft runs it for you, that’s not cloud. You can’t scale up and down, and the system is not automated.
Our system was designed from the ground up to scale in the cloud. It is a true cloud service. When we move our compute machines to the cloud, the appliances that are going to the cloud need to live by those same rules of the cloud. They need to be able to scale up and down dynamically and automatically.
Not just software. Commodity software.
The change is massive. In a way, cloud architecture is subsuming all other architectures. The new OS is the cloud OS. And why not? It’s cheaper to scale, cheaper to run, and more resilient. And Nasuni will be ready as the high-performance, cloud-native file services platform of choice.