Getting IaaS right is your recipe for cloud success

  • Posted on August 16, 2016

Infrastructure as a Service

One of the cloud’s big selling points is the lower cost it delivers to enterprises like yours. So why do some companies see IT costs rise after moving to the cloud—and how can you avoid being one of them?

A key reason for this paradoxical result is that some companies take their traditional views of infrastructure and architecture along with them when they move to the cloud. They think of the cloud as simply another place to host a new version of their existing infrastructure. But with the foundation of cloud savings being the consumption-based model, the traditional approach becomes highly counterproductive in the cloud. When server instances are bought as though they’re needed to populate a datacenter and left running 24 x 7, as opposed to scaled up and down to meet capacity, it eliminates the savings that come from the cloud’s pay-as-you-go model. On top of that, if instances aren’t decommissioned when no longer needed, it leads to a sprawl that’s as unwieldy to manage as it is expensive.

Understandable—and Avoidable

Given the way that many companies begin their journey to the cloud, this problem is understandable, despite being avoidable. Often, management directs that something be moved to the cloud as a trial. IT finds an application that seems a good candidate, and moves it as-is to virtual machine instances in the cloud.

And then the problems begin, because key infrastructure elements just work differently in the cloud. Compounding the error, decisions that are made to support the first app in the cloud may not be the right decisions for subsequent apps and for the enterprise as a whole, leading to redundant and perhaps incompatible infrastructure elements, poor performance, more complexity, and more cost.

Happily, these problems are, as I mentioned, avoidable. Part of the solution lies in identifying which infrastructure elements should be addressed prior to a cloud migration, and which ones that can wait for later.  Even more, which infrastructure services should you continue to build and maintain yourself and which should you simply consume?  The decisions you make will shape the technical cloud foundations for your enterprise and need to be addressed first regardless of whether you go on to adopt Infrastructure-as-a-Service, Platform-as-a-service, Software-as-a-Service, and even Business-Process-as-a-Service.

Do These First, Leave Those for Later

In general, the infrastructure elements to address first are the ones that all services must consume, regardless of their type or purpose, such as identity, networking, security and resource consumption and management. The details of the designs for how you consume other infrastructure elements, such as monitoring, alerting, load balancing, disaster recovery, storage and compute, can be tackled later.

Take identity: In-house identity solutions are often sacred cows due to the level of investment and specialized skills required to maintain the platform.  These investments were necessary but they did not differentiate your core business – with limited exception, customers don’t pay more or buy more of what you’re selling because of your identity solution.  With the introduction of cloud-based identity, such as Azure Active Directory, companies can now choose to consume identity as a service, eliminate the complexity and specialized skills required to support in-house identity solutions, and focus talent and resources on higher value services which can differentiate the business.  Whichever solution you go with, these are enterprise architecture decisions that need to be made thoughtfully up-front and not driven by what is already being used by whichever app you go to the cloud with first.

But Wait, There’s More

Enterprises also have to consider another layer of complexity: Their infrastructures are unlikely to be cloud-only either now or for the foreseeable future. Instead, they’ll likely be hybrid solutions combining the cloud with on-premises elements. Apart from startups, almost every enterprise has solutions that are not technically able to move to the cloud or not cost-effective to move, such as mainframe and other legacy solutions. That means that the infrastructure solution chosen for the cloud must work seamlessly across both parts of the hybrid environment—another challenge, but a solvable one, if the technologies are chosen and deployed correctly.

And Now for the Good News

The good news is that the enterprise that gets its Infrastructure-as-a-Service layer right can go on to gain even greater cost savings and agility from the higher layers of cloud services. As my colleague Heath Whelan points out, public cloud providers are making enormous investments at a scale that most enterprises can’t match on their own, nor do they have to. Enterprises can achieve the tremendous benefits of the cloud simply taking advantage of those public cloud services. And creating a strong infrastructure foundation is your first step.

Avanade Insights Newsletter

Stay up to date with our latest news.

Next steps

Talk to us about how we can bring the power of digital innovation to your business.

Modal window
Share this page