After waiting so long on the sidelines to decide if cloud computing is right for you, you finally decide to jump into the fray. That’s a great first step.
But while you’ve been deliberating, the number of cloud options, models and methods has exploded! Choosing the best one, or ones for your diverse inventory of applications may be – correction, will be —more challenging than you imagined.
Case in point. A recent blog from Cisco, a technology partner of Peak 10’s, was exploring the topic of “So many workloads and so many clouds …” The author makes several great points that are important when selecting which applications should go where and why. Critical success factors are as important for veterans as they are for cloud novices and include:
- Collaboration between IT and LOBs
- Understanding the strategic value of each application, how it’s used and who uses it
- Profiling the performance attributes of individual applications
- Being guided by a corporate governance policy, and driven by a corporate business strategy when implementing cloud services
Before long, the author makes a big leap and skips right to discussing hybrid cloud computing models. Before you go screaming off into the night, hybrid clouds are good things. This is an evolving cloud model of great interest to many.
Hybrid is actually not a thing but a framework that will consist of one overarching computing infrastructure that seamlessly combines and dynamically manages a variety of cloud services and models from multiple providers. That sounds pretty much like what we envisioned way back when that the cloud would eventually become, right?
If you’re just getting your feet wet in cloud computing, understanding and aiming toward a hybrid cloud infrastructure strategy at the start of your journey will put you well ahead of the game in the long run.
Another interesting reference in the Cisco blog is that it couples “build vs. buy” with “risk/reward.” Conducting a risk/reward analysis to assess which cloud option (or no cloud) is the best choice for individual workloads is right on the money, literally.
Some applications and workloads may not perform well in the cloud as they are currently written, which is big reason not to assume all applications will migrate seamlessly. Perhaps they are complex with many interdependencies with other applications, or were not designed with multi-tenancy in mind; oftentimes that can be fixed, but not always. There is cost associated with that.
Then there are applications – custom, legacy, critical beyond measure – that are simply incapable of running in a virtual environment; this is a rebuild scenario. Rebuilding can be a lengthy, costly and disruptive process; however, a thorough risk/reward exercise may come down in favor of it.
However, in this age of “cloud first” directives enforced by many organizations regarding new applications, projects or programs, the build vs. buy paradigm is curious. When you have the opportunity to optimize for cloud computing from the get-go, you will get the performance, security, availability and flexibility you want in an application, with all the attendant benefits associated with cloud computing. When a company has a “cloud first” mantra, the business case for taking any other approach, such as building it yourself, must be justified and presented to corporate management.
So, there’s a lot going on in the cloud industry. Once you think you’ve got a handle on things you lose your grip. But it’s all good, if not a bit daunting.
With so much happening so quickly, it really does take an expert to stay on top of it all. Hooking yourself up with an established cloud service provider that has a range of services and professional skills is the second great step you should take after deciding to give this cloud thing a go. You want the best solution possible for each application, not just an off-the-shelf, lowest-common-denominator service purchased with a credit card swipe.
And most important, you want guidance for growing your business, not just your cloud usage.