Start small (but not toooo small)
So, you’re off and running with your hybrid cloud thing. If you’ve been following along, you’ll understand that you need to know why you (or at least your boss and big boss) are doing it. Hopefully, you understand that you really should act a bit like your team’s therapist, because you’re trying to influence a behavior change and not just implement some new technology. And with the technology in hand, if you focus on the right reasons for your organization (like we did), you can have your cake and eat it too.
The next step in our journey focuses on momentum; not wasting any time in showing real value. In our organization, we have a huge landscape spanning the breadth of technical complexity, and a time span that’s beyond the age of someone entering their first mid-life crisis. With a portfolio of applications numbering in the mid-thousands, it was a daunting figuring out where to start.
We sliced and diced the data and found that we had something interesting. A number — well over 100, in fact — of application teams were running Kubernetes clusters in our public cloud environment. Others were using other technology that we had on the roadmap to sunset. We thought, what if we focused on consolidating these first to move from clusters to an enterprise platform? What would that look like?
Let’s break it down — each additional cluster means overhead: people, finances, more environments to patch. Each cluster means more time to figure out how to set up and operate. Each application team has to contend with basic, repeatable tasks. Our vision of having a single unified experience for the business application teams to give us speed, scale, security, and simplicity would reduce this overhead and help application teams focus more on business innovation and less on the repeatability of managing a platform that provided the core capabilities they needed. For us, this means using Red Hat OpenShift. Managed and integrated correctly, it provides the scale and security isolation that many of these application teams argue is their reason for running their own cluster.
Let’s go back to our fictional example of Sam from our previous article who’s in a similar situation. They believe that hybrid cloud, and implementing a hybrid cloud platform with the right cultural changes in their organization, will help the IT department lead the digital transformation of their company. Sam and their team are figuring out the best way to approach tackling the challenge.
She should consider the following options:
a) First, get organized. Four to six weeks with some consulting firm’s help to develop an overall migration plan is warranted for an initiative of this scope. Then, Sam and her team will have a better idea of what they can accomplish.
b) Narrow the scope to what we’d call low hanging fruit or a part of the application portfolio that is conducive to quickly adopting a hybrid approach. Moving fast and learning as you go is the best approach.
c) Build a platform and advertise to the application teams about how good the platform is. The “build-it-and-they-will-come” approach has worked in the past and so we expect it. /S
Option “c” may have worked a decade ago when application teams were complaining about it taking months to get a VM. But now there are many options that make it easy for them — probably not a good approach for a modern IT operation.
Do you really want to go to the boss and say: “We’ve spent six weeks [and hundreds of thousands of dollars] and have our master hybrid cloud plan ready”? Not if you want to keep your job!
Instead, you may want to say: “In six weeks, we have migrated 12 applications and reduced expenses by $1200.” That’s two applications per week. This is option “b” and identifying the right scope quickly to demonstrate value is key.
Here are two different ways that you might approach identifying the right scope quickly to demonstrate value depending on the state of your organization:
If you have application teams that are already developing container-native applications, look to see how you can consolidate on centrally managed platforms. Benefits will include:
- Increased security through consistent manageability
- Reduced overhead
- Application teams being able to focus more on the business (versus running platforms)
If you don’t have application teams that are doing container development, find an area of the business that is ready for an experiment to develop a new user experience or a digital business process. Benefits will include:
- Demonstrating faster time to market
- Accessing existing data sources through a cloud-native interface
Finding a manageable scope so your organization can learn quickly, demonstrating results that you can measure (more on this later), and starting to operate differently are the keys to success. And you can achieve this, like we did, by starting small and moving fast. From here you increase momentum and start to go big and move faster.
Matt Lyteson is Vice President of CIO Hybrid Cloud Platforms at IBM based in RTP, North Carolina. The above article is personal and does not necessarily represent IBM’s positions, strategies or opinions.
