IT is not a gig for the faint of heart. On one hand, we’re tasked with becoming more agile so we can respond to new business models, competitive threats, and the plethora of new mobile devices. On the other hand, reducing costs is part of our mantra.
This is nothing new for IT; “do more with less” has been the rallying cry for many years. And although virtualization, cloud, and elastic computing present a tantalizing vision for the future, the age old questions of when, how, and how much still resonate with me and my peers.
To whatever extent we embrace these technologies, long-term, most of us are trying to achieve the elusive goal of “on-demand IT”—we’d all like to be able to automatically provision new users and applications with the click of a button, regardless of where those users or applications reside. Frankly, it’s difficult to know if that’s an achievable reality. Whatever the case, it’s clear that cost savings and efficiency gains—in power, personnel, and dollars—are there for the taking with these new (and not so new) technologies.
Getting There from Here
Getting to the point of delivering on-demand IT is an evolutionary process, however. As companies grow and business needs change, yesterday’s simple operation with standalone, dedicated computers has become today’s complex data center. The process of maturing inevitably leads us to consolidate, virtualize, and aggregated servers to cut costs and create “capacity on demand.” Automation tools play an increasingly important role in managing first one, then multiple data centers.
One thing is for sure; as the infrastructure becomes more complex, the challenges along the way just become amplified. Some organizations make it only part of the way to their goals when they realize the cost to manage and orchestrate this complex infrastructure far outweighs the benefits they had hoped to reap. At that point, they simply give up and opt for outsourcing, but they pay a high price for it, losing flexibility, visibility, and control of their applications. Others will keep trying to do it all on their own, but ultimately the business must come first, because downtime is unacceptable.
Outsource Your Infrastructure, Not Your Control
To truly achieve on-demand IT, we need a “middle ground” solution that lets us keep all the control that traditional enterprise data centers provide and blend that with the simplicity that outsourcing provides. For many organizations, this is precisely the allure of cloud computing. In theory, it allows us to combine our own infrastructures with a larger cloud infrastructure. But, is it really that simple? Can we really move applications into the cloud without giving up flexibility, visibility, and control?
The answer is yes, but it requires a new architectural model—one that enables IT to extend its existing infrastructure into the cloud to take advantage of its scalable resources. At F5, we call this the “dynamic services model,” because it brings together all the necessary components for delivering applications and data into a set of reusable services. With these services, IT can take all the context and control that they’ve developed for their data center applications and apply them to the applications that they move into the cloud. This type of architecture is, in fact, essential before IT can comfortably move any applications into the cloud, much less, mission critical applications.
Is Your Cloud Provider “Enterprise-ready”?
If you’re considering moving your applications into the cloud, do your homework. Find a provider that can support your goals for retaining flexibility, visibility, and control. Here are a few questions to ask:
· How do they automate and orchestrate management between the enterprise and the cloud?
· How do they enable access control and ensure that control remains with the enterprise?
· How do they enforce network and application security and demonstrate that enforcement?
· Is the enterprise able to mimic its existing architectures, maintain control, and easily switch from one cloud provider to another?
· What attributes of the end-user and application (device, location, role, performance, availability) do they incorporate into making intelligent traffic management decisions?
· Do they have systems in place to simplify, secure, and optimize virtual machine transfers between the enterprise and the cloud?
Keep asking until you find a cloud provider that can answer these questions to your satisfaction. (Remember, they’re there to support your goals, not the other way around.) Only then will you begin to see that the benefits of cloud computing far outweigh the risks—and that it ultimately has the potential to help you achieve on-demand IT.