When you buy through affiliate links in our content, we may earn a commission at no extra cost to you. Learn 
The cloud has brought an entirely new dimension to the development and implementation of data architectures—and in fact it provides the foundation for many of today’s advanced applications and services.
However, the on-premise data center still plays a key role in most organizations’ business models. That’s because, for one, data centers are substantial investments and shouldn’t be tossed out on a whim. But more importantly, the data center holds a number of key attributes that are difficult and costlier to obtain in the cloud. These attributes are mostly related to control, governance, security and the business sector in which you operate—for example, the financial industry is highly regulated. (Also read: Public Cloud vs. Private Cloud: How to Choose.)
This leaves many organizations stuck between a rock and a hard place: They want to keep their data center for its assets; but there’s an increasing demand for more speed, scale, efficiency and support for entirely new digital processes. Add to this the need for C-suite executives to consider the total cost of ownership (TCO) for both existing and new digital assets, and it’s no wonder many are starting to rethink their data environments from the ground up.
The solution? Transform your data center. But this is far more easily said than done.
Here are three reasons data center transformation initiatives fail, and how to avoid them:
1. Haphazard Cloud Migration
Jason Lamb, cloud solutions specialist at Intel, notes most transformation projects get bogged down in the minutia of infrastructure, believing that simply incorporating the cloud or adopting SaaS architectures will do the trick.
Instead, Lamb says, executives from all departments—on the business and technical sides—need to sit down and define exactly what they need from IT and technology in general. This process should include stakeholder representatives from the wider business because ascertaining wider business plans is crucial for any new business venture. Buy-in from your main stakeholders also makes it easier to secure the budget for an IT project.
Only after they’ve determined their needs can organizations begin to craft solutions to meet them—right now and for the future.
Transformation is not just about infrastructure, however. Too often, says tech blogger Brian McHugh, organizations port legacy applications to the cloud without considering the impact this will have on performance.
The smarter approach is to take a broader view first—to identify how business models and processes need to change to remain competitive. Organizations must recognize they have a greenfield opportunity to create something relevant to today’s digital landscape rather than carry over out-of-date and buggy systems.
With the desired business outcomes pre-defined, the transformation of physical and logical infrastructure—and perhaps many of the business process they support—can proceed along a co-ordinated, streamlined trajectory and stands a far better chance of success. (Also read: How Digital Transformation Can Bring Resilience During Disruptions.)
2. Ill Timing
Time synchronization will be a crucial piece of the data center transformation puzzle, says Markus Lutz, CTO of SiTime Corp., a precision timing system developer. As resources become more distributed across data center, cloud and edge infrastructure, precise timing will allow this collection to function more accurately as an integrated whole. Timing is also vital for critical applications like finance, service fulfilment and security.
In short, timing is everything in network communication: Out-of-sync equipment, whether physical or virtual, can lead to disconnects, integrity and availability issues.
Fortunately, new generations of microelectromechanical systems (MEMs) are transforming resonators and other devices to lower the cost of implementing precise timing across distributed architectures compared to earlier quartz technology. At the same time, they’re improving resilience and reliability and cutting energy consumption by 90%.
3. Clumsy Co-ordination
When everything from finance, HR and supply management to sales, marketing and support is likely to undergo transformation in some respect, it is vital that this process unfolds non-disruptively. Part of that equation will be defining what should remain on-premise, what should go in the cloud and what is best left on the edge.
This can be difficult to determine; but in general the most urgent, customer-facing applications should reside on the edge—this would include things like autonomous vehicles, medical systems data and smart cities functions. The cloud provides the scale and cost benefits of large workload processing and can also be used for archival purposes and high-volume transactions. The local data center should house the most critical data an organization’s knowledge workforce needs on an immediate, regular basis for strategic planning and operational oversight.
In most cases, however, there will be significant crossover between these three layers of infrastructure—so they need to be designed cohesively.
How Cloud is Transforming the Data Center
While it may seem like the cloud has been with us for ages, there is a notable drive toward integration today which represents a fairly sudden and radical shift in what had been a steady evolution in local data resources for more than half a century.
Many modern enterprises—and not just the very large ones—have developed integrated, hybrid architectures consisting of resources on-premises, in the cloud, and even out to the edge. In fact, according to Gartner, this will become the norm as early as 2025 when upwards of 85% of organizations will feature these mixed environments to support increasingly diverse sets of applications.
And although the opportunities in this transition are legion, so are the pitfalls.
Ideally, a transformed data center will be indistinguishable from the cloud—at least operationally. Workloads should gravitate naturally and organically to the most efficient, effective layer of infrastructure. This is being propelled by the rise of intelligent automation, which should, in theory, allow human operators to step away from day-to-day management responsibilities to focus on more strategic matters that provide greater value to the overall data environment. (Also read: Robotic Process Automation: What You Need to Know.)
Conclusion
Ultimately, there is no right or wrong way to go about data center transformation for the new digital age: Every enterprise is different and will require different capabilities from local and distributed architectures. The bottom line? Understanding your organization’s needs and how the data center fits them is the first step to finding the right strategy.
One thing is clear, though: The data center can no longer handle the digital demands of even a small organization on its own. It must integrate with the wider world—and the sooner that happens, the better it will be for the enterprise.
Indeed, many analysts believe integration will be one of the primary drivers of digital transformation. As more and more devices—from wearables to smart thermostats to self-driving cars—become part of the enterprise ecosystem, it will become impossible to keep data siloed within an organization’s network. The Internet of Things (IoT) is creating tremendous opportunities for businesses to revolutionize their industries and create new business models. However, businesses also need to be prepared for the challenges of digital transformation. (Also read: 5 Myths of Digital Transformation.)
