I’ve had my head around, maybe in sometimes, the cloud for quite a few years now. It seems like it’s about all I talk about with customers and colleagues these days. The advent of the cloud has drastically changed everyone’s business and leveled the playing field for enterprises of all sizes. Startups and small enterprises can just go all in and avoid large capital expenditures. Larger enterprises can connect their on-premises investments to burstable cloud infrastructure to support BigData and IOT solutions. All enterprises can shorten rollout times from months to just weeks or even days.
I’m still amazed at the rapid innovation I see on a daily basis. I’m also amazed that every enterprise is not benefiting from it. Most of the push back I get is still around the risk of giving control of your data and assets to a 3rd party company. It takes a lot of trust and some strong contractual agreements with financially backed SLAs. It takes trust that your cloud provider has better physical and logical security than you can provide. They have to have better disaster recovery and business continuity strategies than your team has. They have to let you have your assets back if you choose to switch to a new vendor.
I have worked very closely with Microsoft and their Azure cloud platform for several years. They have implemented tools and platforms for mitigating the aforementioned risks that I would challenge most companies to provide on their own. Logical and physical intrusion detection systems, perimeter security, seismic bracing, geographic data redundancy, etc. Is there still risk? Yes, how do you mitigate that risk? Strategy and planning. With proper strategy and planning, you have an actionable plan to handle those risks. It’s a very true statement, not every business workload is “cloud ready”. Some data is just too sensitive or mission critical to accept that risk. Some of this is because regulations have not caught up. Some is because the cloud platform is not backwards compatible enough. The industry is at a point though in which every type of computing should be measured against the statement, “Why not cloud?”.
Our approach to cloud strategies is very structured and pragmatic. It does not have to be months and months of work. In fact it shouldn’t be. The industry is moving way too fast for that. It is an ongoing iterative approach that will help your organization to continue to move with fast moving cloud enhancements.
The first thing we generally do is take the pulse of an organization with respect to cloud. We measure how the organization feels toward both benefits and risks of cloud. For example, many organizations are not overly worried about being “locked in” by cloud vendors but are very concerned with data leakage. Other organizations think the infinite scalability of the cloud is what will help them out-innovate their competition. While others understand they will have geo-redundant disaster recovery they couldn’t support themselves. Taking this pulse allows us to make the right decisions on which business workloads are the ripest to move and which ones we should just stay away from all together.
The next step in a cloud strategy is to categorize and classify your business workloads. I like to use the word workloads as we can be as vague or specific as we need to be. You have to start somewhere and sometimes staying a little higher level allows us to discover where it is most important to drill in and be more specific. For each business workload, we generally run discovery workshops to answer a couple questions:
- What is the value of moving this workload to the cloud?
- How complex would it be to migrate the workload to the cloud?
- Is it even business or technically feasible for the workload to move to the cloud?
Once we have those answers, we chart each business workload and look for the ones that have the simplest migration path with the most business value. That’s where we start. Get the organization the quickest win.
One of the most common workloads we’ve seen is Dev/Test infrastructure deployment. The recent focus on DevOps is a big driver of this trend. When you can script the creation of servers or server resources, you can solve really big problems in really short timeframes. Unfortunately, many organizations are hamstrung by their lack of infrastructure. If you ask your IT for 60 servers for just a 2 hour period so you can try something, you most likely will get a little pushback. If it is strategic enough, they may say yes, but it will take months. The cloud makes this possible and it will cost much less.
The next part of our strategy is to define guiding principles. These are guardrails that will keep your organization on track in the future in the key areas. It is important to put in place rules on how to select and procure cloud solutions. You also need rules and standards around security and data protection. Most of these guiding principles are similar for every organizations. Some of the unique guiding principles will surface during the discovery process. Having these principles in place will make decision making easier. You don’t want to argue and revisit decisions every time you are moving something to the cloud.
The last part of our strategy is to put a roadmap in place. Often this roadmap is just a picture of what the end state for each of these workloads coupled with initiatives to get there over time. This roadmap should be considered a living document as all businesses have become agile and things will change. Often new cloud services arise that drastically change an organization’s course. That change should always be expected and embraced because it will continue to keep your organization ahead of their competition. The cloud is rapidly changing and your cloud strategy will need to iterate as well to be successful.
One thing to watch out for when putting together the strategy is the political ramifications of the effects of this strategy on the human resources of the organization. Some team members will feel threatened that these cloud technologies will drastically change or even remove their responsibilities. It is important to point out that the cloud allows teams to move from performing commodity tasks to tasks that are strategic to the business. The cloud also provides training opportunities in new areas and tasks that are more interesting and challenging (which in my experience most technologists want).
If you aren’t leveraging the cloud strategically today, you are missing out. Even worse, your competitors are and it will allow them to gain market share. They’ve made the right decisions to create those innovative solutions without huge capital investments. They have redeployed their teams to much more strategic tasks then lower level infrastructure troubleshooting. They are on the road to the cloud success. You may still be able to catch up, but the gap is widening.