Someone, somewhere at your organization is discussing cloud. In strategy meetings someone has said “Let’s move it to the cloud.” At the watercooler or in the breakroom you’ve probably overheard snippets of conversation about micro services, DevOps, cloud strategy, and a host of other cloud-speak. The cloud conversation is everywhere because more companies are moving to the cloud. But they are running into trouble by not defining the expected business outcomes.
The first question you should ask at your company is “What are we trying to accomplish with the cloud?” Too many organizations start investigating cloud options before identifying desired business outcomes.
Without knowing the answer to this first question, it is difficult to define the ecosystem for your cloud. And the second question is “What is the ecosystem you need to enable in your cloud?”
The third question is, how will you get there and what is your strategy? An all too often unasked part of that question is, “do you have the skills, knowledge and resource availability to take this on?”
The harsh truth is that cloud adoption is moving from early adopters to a mainstream competitive advantage. In response, many companies have already started a move to a cloud, whether it be private, public, or hybrid.1
In many organizations, individual groups throughout the business already have some IT workloads in the cloud without an enterprise direction for why they should be there. For example, because development organizations may move faster than infrastructure teams, many companies have already placed some development workloads in the cloud.
Also, a move to the cloud doesn’t necessarily mean a traditional environment is going away anytime soon. It simply means that many organizations will need to determine how to attach a front-end, cloud-enabled service onto their existing IT portfolio. In other words, how can the cloud run a portion of your application workload? Secure, agile, digital enterprises that can use traditional IT and new IT — such as cloud — will be the most effective and successful in the future because they will have the flexibility to cope with changing market conditions in a timely manner.
As an organization, you are going to have to figure out where and how cloud makes sense for you. Is it a technology or a capability you are looking for?
What are the business outcomes that determine success? Where is the business value? Are you simply looking for a way to lower or change your cost model? Or do you want to fundamentally change how your enduser experience occurs from an application standpoint? How are you defining and measuring success?
The business benefits of cloud vary greatly. Some of the benefits can be:Speed to market Application availability Improved resiliency Reacting quickly to changing market expectations Mobility enablement Vendor licensing changes Lower cost Changing cost models Improved client experience
The cloud model you choose will vary based on what business outcomes your organization wants as a whole.
Cloud is no longer a simple virtualization play. It is a transformational effort bundled in one word. It often involves a transformation of how you are a service provider to your business. This transformation can involve complexity and cost.
Cloud is not a single product but an ecosystem of aligned parts to produce a new operational model.
Don’t get me wrong, there are plenty of areas in which moving to a cloud isn’t that complex, such as moving existing workloads — especially development environments — to public cloud providers. However, there are still many areas for existing enterprises which are problematic.
In particular, putting a workload into the cloud doesn’t mean you accomplished your business outcome. You may simply have moved existing challenges into an environment where increased or variable capacity could cost significantly more than it did in your traditional environment. A challenge for many companies is to determine how you should monitor, manage, and react when a portion of workload moves to a cloud model.
So, don’t be the kid in the candy store. Some applications will remain in the traditional environment for the foreseeable future. It doesn’t pay to move them. The questions to validate for each potential cloud investment are:What outcome is required for the organization to move forward? Is cloud the way to get there?
Answering these questions will help you define your cloud solution in alignment with your desired business outcomes.
Overall strategy is critical, whether you are looking at the next two years or farther down the road.
There are three overarching benefits to developing a relevant strategy before starting:Match options to business outcome. Most likely, you are aiming for a hybrid model because it provides many options and different options for different workloads. Accept that organizational impact will occur. By being cognizant of this and planning for it, you are likely to manage organizational change and avoid unnecessary churn and spending. Prevent biting off more than you can chew at one time. Plan the pieces in discrete chunks. Focus on minimum viable product.
A successful cloud strategy defines what success looks like and continuously tests against that definition. It also considers the following:Your legacy environment will be a part of your cloud strategy. Unless you are developing a brand new application, you are going to have to touch the existing environment. You are still going to have to care about how it performs and how it works. You are going to have to care about the inter-relationships between the cloud and your existing environment, even if you place portions of an application in a cloud environment. These relationships will include the existing environment’s security models, development practice, architecture, monitoring, and incident and problem management. They all are going to help you determine the pace of change, the cost of change, and the potential risks. Forge partnerships and learn from early adopters. It is important to develop partnerships and learn from the early cloud adopters. Accept that you are going to need partnerships, both internal and external. Internal because developers and systems administrators are going to have to work together, long-term, not in a short project-based delivery cycle. DevOps requires continual cooperation. Externally, for the first few years, most companies are going to require consultants or managed services to hold their hand if they are doing more than test and development. There is a level of expertise required to run cloud models that you are going to have to develop internally. Once you determine the security and regulatory requirements (which are enterprise requirements sitting on top of your cloud), you are going to have a bit of a hangover. The easiest way to resolve these challenges is to pick partners who’ve done this before and have already run clouds. A “fast fail” culture is critical to cloud success. Many organizations claim to be agile or know how to fail fast but few really do this well. Embracing the “fail fast” culture is essential, starting with the initial cloud deployments, even if they are small-scale. These experiences actually encourage some failures to get people used to going “Hey, that didn’t work!” Coming to this realization that it is okay to fail and learn from your mistakes is difficult for many organizations to practice at a ground level.
Failing fast is not an excuse to impact clients; it is a way of enhancing and accelerating cloud adoption.
In contrast, cultures of blame, shame, and traditional project deployment (financing, waterfall, etc.) impact your speed to implementation and your ability to gain the benefits of cloud.
Break it in Two
To be really successful, we have found that you should break the cloud organization away from the existing IT culture. You should create a culture that is different from the existing organizational hierarchy. It is important to carry over the rigor and discipline; however, the cloud is a long-term change in the operational model. You will task the cloud organization with becoming the leaders of the operations and technologies that are needed to get to the cloud. Defining the cloud team is essentially defining the early adopters for your cloud.
Factor in Failure
In some cases, there is a lack of awareness of the pitfalls associated with a move to the cloud or a lack of adequate compensation for those mistakes. Ensure that failure is always factored into the equation and that resilience is built in; this is a cloud architecture design best practice. For example, companies need to understand that components will fail and there needs to be a design that prevents such failures from impacting service delivery. This is a difficult concept for many people to grasp when they first start developing cloud deployments.
Start with One Application
So, now that you have resolved your cultural challenges, what next? Normally, we suggest you identify a specific application to migrate to the cloud. Why? Because you are going to have to change your operational processes along with the culture and incident management. If you align to an application, this will help with change and incident management and it will drive success in the cloud.
Use A Scorecard
It can be extremely beneficial to scorecard a selection of the business applications and evaluate them both on business value and complexity to run in the cloud. Then, it is important to make a selection on which one will drive business value but allow rapid progress and add value to the organization. Low-complexity, high-value applications are best to start with. Some organizations move these applications — as proof of concepts on test clouds — to prove the concept works before they move forward with a full cloud strategy.
Realign Your Security Model
Normally, we advise that security come first, in front of IT operations management (ITOM) and IT service management (ITSM) models. However, without understanding the specific details of your cloud, it is difficult to have a valid security model. Therefore, the last —and most crucial step — is to realign your security model.
Due to the nature of how clouds work, a good deal of the existing solutions used to define security aren’t as effective. Instead, you are going to have to evaluate your controls and then see how you can meet that requirement in a cloud model. That’s going to require new techniques, tools and, in some cases, re-evaluating those controls.
Who will Execute on the Strategy?
Because of the significant changes in mindset and technology involved in enabling a cloud ecosystem, it is critical to determine if you have the skills, knowledge, and resources available before you start.
Building a cloud is not that hard for companies with the technical chops. What is harder is getting your application to the cloud. It is even harder to run a cloud two years later. Harder still is delivering on your desired business outcomes consistently.
If you really want drive business value, organizations need to change how applications are designed and operate. An organization could end up creating a future backlog of redo work for itself or not achieve its goals if applications are not designed without considering session state, microservice design, and mobility. For example, changing processes to a continuous integration and continuous delivery pipeline. This move which allows the applications to do smaller, incremental, faster changes needs to be aligned to both the cloud and application via automation.
In my experience, actually having the application in the cloud is not as costly, and as scary, as many companies perceive. What is complicated is the other organizational processes surrounding the application that is now in the cloud.
The move to the cloud is neither simple nor cheap, but the payoffs can be substantial, and it is increasingly a necessity to retain competitive advantage. Taking the right steps, in the right order, beginning with the right conversations, can mitigate risks and enhance benefits.
What conversation is your organization ready to have? And when is your organization ready to start? Because someone is already having the conversation, and someone is already starting.
. Hybrid cloud is defined here as a cloud environment which utilizes a mix of on-premises, off premise, private and third-party, public cloud services with orchestration between the environments.