With today’s businesses increasingly relying on public cloud services such as AWS, Microsoft Azure or Google Cloud Platform, there is an inevitable transition period as they migrate legacy on-premises IT infrastructures. Most organizations opt for the hybrid model to sail through the transition. Michael Bathon, vice president of cloud services at Rimini Street, however, warns against using a long-term hybrid IT model and discusses three major drawbacks.
Since there isn’t a mystical button you can push to somehow migrate everything within your data centers to the cloud in an instant, organizations need to ask themselves one key question: how long should we utilize a hybrid model when it comes to our IT infrastructure?
The short answer? Not long.
Yes, a transitional hybrid setup is indeed necessary during the migration process. In most cases, however, organizations are better off going “all in” on the cloud to the best of their ability and following a business roadmap that steadily gets them there over a two- to three-year period.
Why Are Companies Migrating to the Cloud?
For various reasons, talent is an essential driver of cloud migration. First, as older on-premises hardware ages, the number of people with the skills to maintain it decreases. Professionals frequently leave or change occupations at this level, and there aren’t many younger people with experience working with antiquated IBM equipment, for example. Long-term organizational know-how — “tribal knowledge” — is precious, but it can be costly to preserve.
At the same time, organizations have a growing desire for cloud IT infrastructure because that’s where the bulk of talented young professionals are headed. The aim is to develop, refine and retain a solid foundation of talented individuals over time, becoming much more challenging with infrastructures that live largely in physical data centers.
When it comes to on-premises versus cloud IT, different skill sets are required. On-premises hardware, for example, differs from cloud hardware at the network, database, and application levels, creating a need for individuals familiar with the various intricacies.
Three Common Pitfalls of a Long-Term Hybrid Model
Utilizing a hybrid model forever is technically possible, but budget restraints make it impractical for most organizations. There are three main reasons why attempting to do so could be an unfortunate business decision:
- Perennial costs hit critical mass: The physical footprint supporting an organization’s on-premises architecture — which was most likely created when it made financial sense — can eventually stop producing ROI. I equate this to living in (and paying for) a house with way more space than you need. It may look nice, but paying for the unnecessary space is inefficient in the long run.
- Potentially difficult in-office dynamics: If a company has both cloud and non-cloud crews, it’s effectively telling its employees that they’re either on the varsity or JV teams. I’ve seen this type of situation cause issues within a team because everyone wants to be a part of the cloud team, which is understandable. If business conditions dictate that the cloud is the way of the future, what will the non-cloud team members do in three or five years? Businesses must be cognizant of the long-term consequences of this type of circumstance.
- Additional administrative support is necessary: Professional systems administrators must support both on-premises and cloud infrastructures when utilizing a hybrid approach. Tasks including patching, monitoring, failover, backup, and restores are the responsibilities of these teams, and they require both more labor and more knowledge.
The Real Key To It All: People
In today’s landscape, the IT skills shortage is a real phenomenon. This means that organizations need to design their IT infrastructures accordingly. Just about every modern business needs coverage 24 hours a day, seven days a week, but they can’t all employ enough engineers to keep this a reality for 365 days each year.
Yes, some businesses have built small but reliable IT teams, but these teams almost always hit a point where burnout becomes a factor. A certain level of anxiety comes with knowing that it’s not a matter of if an incident occurs but when, and this can grind innovation to a halt.
Still, knowing that a transitional period of hybrid IT is almost always necessary, there are three keys to making cloud migration as efficient as possible:
- Get all your people on the same page. From top to bottom, organizations need all of their people in lockstep and moving toward the same goal. Everyone within the company should understand why a move to the cloud is good for business, and they should know their role in doing so.
- Analyze the complete total cost of ownership (TCO). The online TCO calculators provided by the respective public cloud providers can only tell you so much. The true costs of a cloud migration can vary significantly from business to business, and figuring out what this means for your individual organization requires a much deeper dive.
- Create a tangible two- to three-year plan. Build a business roadmap that includes concrete and incremental goals that steadily move the organization toward where it wants to be.
Phasing IT Out
Prioritizing which systems to migrate first is also an important consideration. Typically, moving both internal-facing and customer-facing systems can deliver significant ROI, but it’s also important not to forget other backend network devices or file systems. Neglecting these for too long could become a costly problem at some point down the road.
Cloud migration is not a one-size-fits-all proposition. Every situation is unique, and how your organization ultimately gets there could be different compared to a peer. With that said, one thing almost always rings true: limiting the length of time you spend operating under a hybrid model gives you the highest chance of success virtually every time.