The cloud has revolutionized information technology and the data center in many ways. It has eased the burden on IT. It has allowed companies to backup data with confidence. If you ask anyone on the business side, however, the first thing they’re likely to mention is cost. Is cloud more or less expensive than on premise? How do I quantify the benefit of scale?
Over the past three years we have seen the cost of cloud decline significantly. In 2018, there is no reason to expect this trend to stop, and the other great trend of increasing performance should also stay strong. This is a boon for businesses of all sizes, and should allow some other interesting trends to develop. In this article, we’ll look at three exciting projections for the coming year that will come out of this increasing performance-to-price relationship, and what they mean for you.
Consumption-based Models Thrive
Overprovisioning has always been a problem for on-premise infrastructure. Companies buy and provision compute with an understanding of how the compute is or will be used and what the computational gains between generations of servers will be. This leads to having more compute provisioned than you actually need which impacts space, maintenance and licensing costs.
What is surprising is that this same bad habit is making its way into cloud planning. When mapping on-premise compute to cloud, many organizations are using a manual, static process of comparing what they have provisioned on premise and finding a direct match from within the cloud catalog of options. The ideal process here is to evaluate provisioned compute including age of hardware, usage and software licensing types to choose the best cloud compute platform and size for the job. This of course is hard to do manually, but there are analytics platforms that can help. By doing this, overprovisioning vanishes and you pay for what you need while eliminating wasted expenses. The savings this can create are not insignificant. And, according to recent research by TSO Logic, it can be up to 36 percent. With cost considerations still driving IT to a large degree, continued rapid adoption of consumption-based models is a virtual sure thing.
Software Licenses Matter When Moving to Public Cloud
Understanding which licenses (like Microsoft) can move to the cloud is nuanced, however it’s important to understand when you have to repurchase and when you can transfer – and there is often more than one way to do it.
Here’s the good news: in the modern public cloud, many software licenses are very much portable. And, in many cases, the same Microsoft licenses and enterprise agreement discounts you’re using in your on-premise data center can be ported to cloud providers. Based on our analytics the savings are meaningful and represent up to an 88 percent savings when moving existing licenses versus purchasing the same license from the cloud provider.
Applications Drive Stickiness
From an enterprise perspective, the cloud’s main use case over the last 12 months has been largely in infrastructure migration (IaaS). As the infrastructure lands, the focus then turns to modernization which includes offloading the maintenance of that infrastructure. Platform-as-a-service (PaaS) models are further freeing IT from those rote, daily tasks, but this time it’s on the application side.
The use cases are many and varied. It may start simple with a move to a fully managed database stack or it might take a more holistic form and start using server less functions. The benefits of going to the cloud for applications go far beyond cost and IT resources, however. Just as each application has different requirements — for compute, for security, etc. — each cloud provider has different areas they excel in, so companies can now choose the provider that best fits their needs.
As applications start taking advantage of cloud native servers they are able to build and scale faster, however it will become more difficult over time to shift away from the chosen cloud platform. This makes the cloud for applications “sticky” and makes that initial research on where to move your applications critically important.
Continuous Migration is the New Normal
While the cloud simplifies IT, the actual migration can be lengthy, at least the first time around. You have to figure out what should be moved first, plan, test, and then actually execute the migration. And that’s before including factors like creating a business case for the cloud in the first place. Because of this complexity, it’s common for organizations to revisit cloud options less than they should.
How often should organizations evaluate their cloud options? Almost constantly. The options cloud providers offer change virtually every day. As technology improves and prices fall, consistent analyzing of your workloads and evaluating options will deliver real cost savings and performance improvements. So, get a documented process, continuously evaluate the options and your own workloads, and migrate whenever there is a clear business case. The savings will be well worth the effort.
These are far from the only benefits companies will see from the cloud’s increasing performance options and decreasing price points. As the technology matures and new technologies emerge, the new use cases for the cloud will be exciting to witness — and it will have profound effects on all aspects of business.
