Hybrid Clouds: The Long Road Ahead

When it comes to enterprise IT these days, just about everything involves some form of hybrid cloud computing. The challenge is that there are so many forms of hybrid clouds that many IT leaders don’t realize just how extended a journey their organization may be on.

The typical IT organization usually embraces cloud computing first with a few software-as-a-service (SaaS) applications. In that regard, the hybrid cloud scenario that emerges is relatively simple: IT leaders need to find ways to share data between existing on-premise applications and SaaS applications that are most often servicing the needs of a specific department or line of business.

IT leaders also find themselves trying to manage infrastructure-as-a-service (IaaS) environments. IT usually starts out with a few developers taking advantage of platforms such as Amazon Web Services to build and test applications.

Eventually, some of those applications wind up being deployed in production. The moment of hybrid cloud truth comes when the organization decides to deploy production applications on a public cloud. Naturally, the primary reason to take that path is to create a more agile IT environment that is more responsive to the changing needs of the business.

A case in point is Giant Eagle, which operates a network of retail outlets selling everything from groceries to gasoline across the United States. Jeremy Gill, senior director of technology infrastructure at company, says Giant Eagle is making a strategic shift to employ IBM SoftLayer as a provider of an IaaS platform on which it plans to deploy a broad range of applications.

“We’re moving all of our virtual infrastructure to a cloud-first model,” Gill says. “We want to be able to stand up infrastructure as needed.”

Like many IT organizations, Giant Eagle is challenged by the fact that it takes the internal IT organization weeks to provision IT resources, but it may take only a few minutes to provision a virtual machine. But all the network and storage resources that need to be configured for that virtual machine create a complex endeavor.

Gill says employing IBM SoftLayer gives Giant Eagle the flexibility to deploy a virtual or physical server as needed. It also provides Giant Eagle, which today operates two data centers, with a lot more flexibility in terms of where application workloads can be hosted in the U.S.

Brokering Services Between Clouds and on-Premise

While Giant Eagle is making a strategic decision to embrace the cloud first, Gill acknowledges that making that shift is an extended process. Existing applications are not going to move into the cloud over night, and he’s well aware that the tools needed to broker IT services between public clouds and on-premise environments are still in their infancy.

But that’s only the beginning of the complexity that hybrid cloud computing will entail. As more production application workloads move into the cloud, the need for robust platform-as-a-service (PaaS) technologies becomes more acute. Before most IT organizations know it, they will be managing multiple PaaS environments running across public and private clouds.

For example, one application might be running on an instance of a Cloud Foundry PaaS running in an environment managed by a hosting provider. Another set of applications may be running on the Heroku PaaS environment operated by Salesforce.com. Marry that up with all the investments IT organizations have in legacy applications, and it quickly becomes apparent that managing IT in the years ahead is about to become both simpler and more complex.