With increased flexibility, cost savings, performance, and speed, more companies are migrating their in-house infrastructure and application workloads to the cloud. Additional benefits that make this option viable include data security, the likelihood that profitability will sky-rocket, and greater employee satisfaction from those who prefer this newer, smarter way of operating.
Despite the benefits, cloud migration does not happen automatically or magically. Successfully moving applications without causing interruptions in operability can be realized, but first, firms must construct a well-thought-out planning phase. Where planning is absent, there is a tendency to approach the transition to the cloud with the same mindset as moving to a new home, “Pack it all up and hurry! We have a deadline! We will deal with it later!” That process is likely to bring about unforeseen consequences to the detriment of employees and the organization itself.
Navigate around these common mistakes
Often, companies are somewhere on the continuum as they make mistakes that hover between the organizational norm and the end goal of operating successfully in the cloud. Fortunately, there are ways to avoid these common missteps. Here’s how:
- Recognize a cloud-first strategy is not a cloud-only directive. A common misconception is that moving to the cloud requires the elimination of a legacy system in favor of total and immediate migration to the cloud. Rather, a phased approach and carefully developed strategy are best. While a lift-and-shift method may work for some organizations, most often it is beneficial to retain some aspects of the existing system—at least for the time being. In the planning phase, a key consideration is identifying which applications should be moved first. The answer is those that will cause the least interruption to the business operations and its customers. The worst-case scenario is a hasty transition that causes unstable operations.
- Think about and plan for what applications to migrate. Neglecting the careful analysis of the current system prior to cloud migration is a detriment to strategy development. In deciding what operations to migrate and the timing in which to do so, the efficiencies and total costs of ownership of the current system must be well understood. The harsh reality is that not every application is still serving the organization in an efficient manner. Migrating data center capabilities does not have to be a “no application left behind” tactic. Another way to look at it is to take the time to evaluate the R’s—retain, re-platform, re-architect, replace and retire as they are the foundations of migration strategy decisions:
- Retain. Do applications exist that are irreplaceable? Make sure to retain those that create a competitive advantage.
- Re-platform. Assess the degree of satisfaction with an application. Maybe it could be slightly modified for performance, but the decision to do so will require diligent analysis of the intended functionality to see efficiencies going forward.
- Re-architect. Do applications exist that are not functioning ideally or are incapable of easily upgrading in terms of capacity or speed? How much has been spent in the past to upgrade a non-dynamic application existing in a very dynamic environment? Rather than migration, the long-term benefit may exist in re-architecture by replacing certain components with cloud native capabilities or re-designing the entire application to be fully cloud native.
- Replace. Widely used “basic” applications can be easily replaced by the service provider when migrating to the cloud by leveraging Software-as-a-service (SaaS) or, Platform-as-a-service (PaaS) alternatives, resulting in initial and future cost savings. Mistaking an application for a one-size-fits-all solution; however, will have consequences for operations if the application is unique to the organization. This is especially true if the selected cloud service provider does not have the ability to replace the existing application with the same level of functionality in the cloud.
- Retire. What elements of the application stack are not in use, archaic, or represent cost centers that do not contribute to profitability? When leveraging something new, do not make the mistake of passing on the opportunity to dispose of non-performing applications. Seek these out and retire them.
- Consider the big picture – risk versus innovation: Ask where does the cloud intersect with the overall business strategy? Risk is an unavoidable counterpart to innovation initiatives. Cloud migration is no exception. It is important to have an established point where cloud service capabilities intersect with an organization’s strategic business needs. Mistakes in proper security, disaster recovery, and compliance are preventable if recognized. The mistake of assuming that cloud service providers or CSPs universally include solutions for these risks can result in the vulnerability of great consequence. Here are common assumptions, followed by the realities:
- Risk assumption #1. Security is increased and guaranteed when operating in the cloud. The reality is security is a joint responsibility that requires diligence and follow-through from the organization as well as the service provider. A firewall being offered by a cloud service provider offers no security if the organization does not use it.
- Risk assumption #2. Operating in the cloud mitigates the risk of human error and physical incapacitation of a single, in-house data system thus, disaster recovery is not the risk it used to be. The reality is that while it is a relief to eliminate the possibility of human error and physical damage to a single data center, the cloud cannot eliminate the risk of natural disasters, or other service disrupting events. The upside? It goes above and beyond by providing options like fault domains, availability domains, and multiple regions. It is up to the organization to design and architect the cloud environment intelligently.
- Risk assumption #3: Migrating to the cloud means that the organization is operating in compliance because the cloud is all-knowing. The reality is compliance needs to be specific to the overall structure of the organization it is serving. Failure to meet requirements set forth by PCI-DSS, SOX, or HIPAA for example, would be devastating. While CSPs can provide options and technologies to ensure compliance- data redaction or encryption capabilities, for example- it is up to the organization itself to ensure appropriate configuration of these capabilities.
- Cloud service providers do not offer employee retention as a service. Peter Drucker once said, “The greatest danger in times of turbulence is not the turbulence—it is to act with yesterday’s logic.” Migration to the cloud will register in many employees’ minds as extremely turbulent, job-threatening, and worst of all, unexpected. Organizational change is not to be approached lightly with a project of this scale. If employees are shocked to learn of this move, they only have yesterday’s logic to reflect on. A top-down approach is necessary and a careful strategy to align employees with the goals of this endeavor will prevent the potential for total chaos during the transition. This strategy should include asking employees for feedback and participation, training for weathering the migration, and clear expectations of roles and opportunities once migration to the cloud has taken place.
Managing change: thoughtful leadership for the win
While an organization can survive without a physical data center, it cannot survive without a competitive strategy and employees willing to see it through. Technology innovations can be turbulent undertakings or golden opportunities. Where an organization lands on this dichotomous scale is determined by a planning phase that addresses all impacts, immediate and future, of implementing cloud migration. This is a time for leadership to steer the proverbial ship into uncharted waters with carefully plotted navigation. Risk is an inevitable counterpart to innovation. In mitigating risk, planning and strategy are an organization’s best prevention to common mistakes when migrating to the cloud.
Credit: This article was first published in IEEE Computer Society