Even though the world is a decade into the cloud era, organizations of all shapes and sizes are still considering what moves, if any, they want to make with regard to the cloud. So, it makes sense to look at what the cloud offers. Or more specifically, why organizations are making the move to public cloud. Let’s start by taking a look at the existing on-premise model to identify potential pain points.
With a traditional data center procurement model, you buy everything up front — a serious capital expenditure — use it for as long as you can, and then throw it all away and begin anew with then-current equipment. If something unexpected happens in the middle of the cycle, you’re left scrambling to figure out the budget and operational impact.
What about the process for buying more storage capacity in your current data center? You likely have a series of approval processes you have to go through after which a purchase order is cut and then you wait for the new equipment to arrive on your dock. Next, you spend some time racking and cabling the new storage and you then go through whatever deployment process is required by the vendor. The setup isn’t likely to be all that involved from a technical perspective, but it can take weeks to months to complete because of the nature of decision-making and the realities of the shipping process.
If you needed that capacity right now, you’d be in a bind, wouldn’t you?
For argument's sake, Nutanix’s core business has greatly simplified this experience with our hyperconverged infrastructure (HCI) solutions. Hyperconverged infrastructure streamlines the deployment, management and scaling of data center resources by combining x86-based server and storage resources with intelligent software in a turnkey software-defined solution. Separate servers, storage networks and storage arrays can be replaced with a single hyperconverged infrastructure solution to create an agile data center that easily scales with your business.
However, when compared to the traditional dispersed data center, a cloud model is exponentially more agile and dynamic. With the public cloud, the procurement and deployment process consists of logging in to the cloud provider’s management portal and clicking a few buttons. Within seconds, you have that extra storage that you so desperately need. Of course, your next monthly invoice will reflect the fact that you just asked for more services. And that’s why moving to the cloud is so attractive.
From an economics perspective, cloud allows you to adopt a pay-as-you-go mentality. You buy what you need today. When you need more, you buy more. With the cloud, you have no capital costs directly applied by the cloud service provider. Everything is an operational expense. Rather than driving a truckload of money to an enterprise IT vendor, you’re essentially paying a rent check each month to a cloud service provider.
Cloud Adoption Doesn’t Mean You Abandon the Data Center
Yes it is true - data centers have traditionally been complex beasts to wrangle. CIOs have had to build teams to manage each component and ensure those teams support a complete operating environment. A daunting undertaking for most in that position indeed. Hence the rapid growth of cloud computing. However, I do want to ensure that you have an accurate view of how this growth is manifesting itself within organizations.
You may be thinking that a cloud adoption process involves your company sending checks to a cloud service provider and then taking sledgehammers into the data center and “carefully removing” the equipment you just stopped using. Reality is a tad different. With some exceptions, organizations aren’t throwing everything into the public cloud. They are retaining on-premises data centers that operate in conjunction with their public cloud deployments, creating what has become known as a hybrid cloud. Public cloud adoption has certainly had an impact on enterprise IT vendors, but the fact is that companies are still building data centers.
Private Cloud: Your Other Friend in the Data Center
In all this talk about the public cloud, you can easily forget that there is a flip side to the cloud coin — private clouds. We absolutely must address this since we’ve just established that your data center will still be relevant. In the simplest terms, private clouds are on-premises data centers that operate similarly to public clouds. The public cloud can be a compelling location in which to operate workloads, but it can have serious downsides as well. Some applications simply can’t run in the public cloud or you may have discovered that public cloud economic benefits don’t hold true for your particular use case.
However, you still may be able to have a public cloud-like experience. It takes rethinking how you operate the on-premises environment. You can’t simply virtualize everything and then declare, “I am now a private cloud!” You need a management and orchestration layer that brings to your environment the kinds of management capabilities that are inherent in public cloud.
By making the investment necessary to bring your data center to private cloud standards, you can enjoy some of the benefits of the public cloud while retaining the control you currently have over your infrastructure.
Your private cloud environment is increasingly likely to be part of an overall suite of cloud services that you deploy. Many organizations are augmenting their private cloud environments with services from multiple cloud providers, thus creating multi-cloud environments. Organizations leveraging multi-cloud services choose which services they want from which cloud, providing their organizations with a great deal of flexibility when it comes to new service deployment.
What Organizations Are Doing in the Cloud
To be honest, every organization has their own unique set of priorities to tackle leveraging the cloud. However, the initial step IT teams usually make is to modernize existing, on-premise delivered services. And something we’re seeing move more often to the cloud are desktops and apps. When you install an application the traditional way, it installs a bunch of hooks into the operating system to make various aspects of the application work. However, the desktop environment is often not static, meaning that a lot of variance exists in the hardware, operating systems, and applications in use across an entire organization’s desktop fleet. Unfortunately, some applications simply won’t work in some environments, whether it’s because of a hardware issue, an oper-ating system incompatibility, or a conflict with another installed application.
Desktop fleet variance creates an inconsistent environment into which to deploy applications, significantly increasing the challenges associated with managing these systems. This is where application virtualization comes in.
When it comes to virtualizing desktops and applications, virtual desktop infrastructure, or VDI, has been the dominant force. Essentially you virtualize desktops and run them from the data center. But an emerging technology in this space is newcomer desktop-as-a-service, or DaaS. DaaS offers a pay as you go model that typically operates in the cloud. We’re actually advocates of both models but which to choose ultimately depends on a number of variables. With that said, to help you with your evaluation process, I’d like to offer you a free download of our Desktop-as-a-Service (DaaS) for Dummies Ebook. Within the ebook, you’ll find guidance comparing and contrasting cloud options, as well as cloud economics. The ebook also discusses the reasons for adopting desktops in the cloud.