The cloud migration plan that many companies follow largely consists of a single step: migrate. They pick an application that they think they could manage more effectively in the public cloud, shift workloads to a cloud provider’s environment and then see how things go.
The problem is that it’s easy to migrate resources to the cloud, but it’s not so easy to do it right. And if organizations take an ad hoc, piecemeal approach to cloud migration, they risk creating inefficiencies, unnecessary costs and cloud sprawl.
To facilitate a successful cloud migration, organizations should incorporate five factors into their plans:
Many companies attempt to incorporate the cloud into their organizational culture only after they’ve already moved resources. But, in my experience, companies that proactively build a culture and awareness around the people, processes and technology that will fuel a cloud initiative are typically more effective at implementing their visions. When organizations fail to do this work upfront, IT staffers may worry that their jobs are effectively being outsourced, causing them to drag their feet. But when companies take the time to listen to their employees, train them for new roles and build buy-in, workers are often able to provide more value after a cloud migration than they were before.
Before moving resources to the cloud, organizations need to engage in application mapping to figure out which interdependencies various applications have, and then incorporate those interdependencies into their plans. During this stage, IT teams may discover that they need to do more architecting, replatforming or rewriting than they initially planned. Or, they may decide to focus on the low-hanging fruit at first, migrating only easy-to-shift applications and infrastructure for the moment, and then moving on to more difficult projects at a later date. Either way, they’ll likely avoid the nasty surprises that can pop up when cloud migration teams fail to do their homework.
For some applications, such as email or file sharing, moving to a Software as a Service (SaaS) solution may make the most sense. For others, Infrastructure as a Service or Platform as a Service will be a better fit. Organizations should think carefully about which resources should go to which model — before they begin their migrations.
While public cloud resources are largely commoditized, organizations should seek to understand the strengths and potential drawbacks of major cloud providers before moving workloads to any of them. For instance, Microsoft Azure, Amazon Web Services and Google Cloud Platform all offer different services and capabilities around features such as analytics, Internet of Things and niche use cases. Typically, we recommend that companies start by moving resources to only one of the major hyperscalers and then decide later if they want to move to a multicloud model.
A cloud deployment isn’t just about IT resources; organizations also need to think through and establish governance, security, compliance and finance policies around their cloud environments. It’s important to implement rules that adequately protect data and applications, without being so restrictive that innovation is inhibited. Organizations that haven’t gone through this stage before will likely want to rely on the expertise of a trusted partner such as CDW. A partner can help establish appropriate frameworks around cloud deployments, provide accurate cost comparisons, suggest best practices for migration and help organization to avoid common pitfalls — ensuring that a cloud initiative meets or exceeds the company’s goals.
This blog post brought to you by: