How long can On-Prem and the Cloud coexist?
The evolution from on-premises to public cloud inevitably sees companies in an intermediary phase, where IT infrastructure sits in a hybrid state while priorities are aligned and investment builds. Unless you're a start-up going all-in with cloud from day one, there's no way to leave your in-house data centres behind overnight.
That begs the question: how long should the transition take?
Organisations that have committed to shifting all their infrastructure to the cloud need to equally commit to a plan that takes them there quickly, pragmatically, and incrementally over two to three years.
Living in a prolonged lull – where the transition takes five or more years – hurts operations and the bottom line. Unlimited budgets aren't a reality, and having a hybrid infrastructure play is detrimental for three main reasons:
- It demands expansive administrative support. A hybrid model requires professional systems administrators supporting on-premises and in the cloud. These experts are responsible for patches, monitoring, failover, backup, and restores. This is far more than some extra work bolted onto their existing workloads – it's extra knowledge and more tools to get the job done, all of which makes migrating more expensive than it needs to be.
- Hard costs reach a tipping point. It's no secret physical infrastructure has a shelf life, and becomes a budget burden than a value generator once it has reached its five or seven-year life expectancy. Holding on to it means paying for kit you don't use.
- Policy conflicts. Having both a cloud team and a non-cloud team runs the risk of resentment as staff are effectively told to pick a side and serve their best interests. Companies going cloud infrastructure first should expect everyone to back that horse. This leaves IT workers managing on-premises in a predicament as they worry about both what they will have left to do once the on-site data centre is gone, and whether they will be left behind from a skills point of view.
It's all about your people
The skills needed across on-premises and cloud infrastructure can vary greatly. For example, the management and support of on-premises hardware and network devices typically use different toolsets versus the cloud. This includes monitoring, performance management, and implementation support. And it may be more than a difference in terminology in how these tools work, as infrastructure-as-a-service (IaaS) management and security tools are typically quite different in function and use compared to on-premises tools.
That makes it all the more important leaders consider what it will take to support their on-premises and IaaS based on the talent that's available. Considering all businesses – but particularly enterprises – now require around the clock support, how many engineers will be needed to take care of all systems year-round?
Historically, organisations leaned on experts being available at all times, including for those dreaded midnight calls that would see them patching overnight to avoid a cyber security vulnerability. So much for work life balance. Planning activities is always an issue because you know you're accountable; should something happen, you lose your weekend. This ultimately affects innovation, as it's unreasoanble to expect employees to be permanently on call, and then also have enough in the tank to implement that next technology solution that drives the company forward.
Balancing resource availability and the challenge of an inevitable hybrid infrastructure model means the focus needs to be on making the transition as efficient and cost effective as possible. That starts with:
- Pulling all stakeholders into the discussion. IT is no longer a backroom silo. Technical leaders must team with business heads and CFOs to strategise and determine why each step of a cloud migration makes business sense.
- Perform a full total cost analysis. Figuring out every cost involved in a cloud migration takes much more diligence than the very basic online calculators provided by major cloud providers. There are many variables, including difficult to predict circumstances, that can quickly drain budgets and make the migration far more challenging.
- Build, and stick to, a three-year roadmap. Create a plan to migrate to the cloud incrementally based on business priorities, and stick to it. Many organisations can get distracted by forgetting the underlying pillars of their cloud strategy, which compromises plans from moving forward.
While there won't be a silver bullet that works for every business, these considerations provide a foundation to fuel progress and avoid a stagnant, hybrid state with extensive consequences. The road to public cloud will look different for everyone, but the longer you spend on a hybrid highway, the more funds you will burn – on both the petrol that keeps you going, and the inevitable maintenance costs that compound as your vehicle ages.
