DCD Opinions: Is the key to the future of next generation data center delivery in our past?

Why yesterday’s lessons could pave the way for tomorrow’s progress.

DCD Opinions speak to Barry O’Sullivan, Kirby Group Engineering.

“Life would be infinitely happier if we could only be born at the age of eighty and gradually approach eighteen”, said the great American writer Mark Twain, concerning knowledge being the very close relative of experience.

How many of us would never approach a task in the same way if we knew what lay ahead before an outcome?

And so, we find ourselves as an industry looking to the next ten years of data center design and delivery and trying to unpick what the future holds for our industry, one that has been the fastest-growing of any demand-based business in the last ten years.

But maybe, the key to understanding our future lies in our past, and hindsight is 20/20 after all. So, what lessons have we learned along the way, and how can we ensure that we put in place best in class process and planning in the future, so that we don’t repeat our mistakes.

My adventure in the data center industry began ten years ago when I was given a chance to work as an engineering manager on a project code-named ‘Project Edge’ in Dublin. At the time, it was an industry first for one of the world’s top data providers.

The hours were long, the pace was fast, and the technical challenges were many, but it was exhilarating in those heady days. I can’t quite put my finger on the reasons why that project has left such a positive impact on me, possibly because there were so many firsts.

It was my first-time using BIM, which was in its infancy, and, in some respects, it still is. Its use was largely restricted to primary plant, equipment, and service routes. Builders’ works were produced in 2D CAD, based on 2D outputs from the 3D model.

The design followed the traditional design process, whereby the design model was a ‘Design Intent’ model where generic libraries were used for plant and equipment, as many manufacturers did not have digital families available.

All parties involved appeared to be in the same boat concerning the challenges we were collectively facing working with a crossover from 3D to 2D in the coordination process. The lack of 3D vendor information contributed to this quasi 3D/2D coordination scenario.

In those burgeoning days when hyperscale data centre design and delivery was in its infancy, there were certain things that we would have done differently or, in retrospect, not at all if we only had the foresight.

As a result, the coordination process was relatively long, with significant coordination changes required to reflect the vendor data. In addition, the term ‘digital twin’ was only used by NASA at this time to improve physical model simulation for spacecraft. It would be some time before it would be used in the context of data centre construction.

However, in those burgeoning days when hyperscale data centre design and delivery was in its infancy, there were certain things that we would have done differently or, in retrospect, not at all if we only had the foresight.

“Get it right the first time” should be the warrior cry for data center construction leaders for the foreseeable future. But, back then, it was anything but.

‘BIM 3D modelling’, was unleashed upon the industry to modernize and revolutionize the design for the construction process.

The reality was that most were not ready to use it, and while the potential advantages were apparent to all, its adoption was not.

As a result, the industry proceeded down a path of using new tools based on a 20th-century design for the construction process that was not aligned and lacked the investment, resources, competency, and, arguably, willingness to make it work.

The result was a highly inefficient design for the construction process, where increasing resource levels offset the inefficiencies, often spilling over into the project’s construction phase.

As an industry, getting it right the first time was undoubtedly going to be a challenge.

I want to be able to say that ten years on, it’s all now sorted. But, unfortunately, not so, and while, undoubtedly, there have been significant improvements in the industry to improve the project delivery process, there remains some substantial challenges and inefficiencies yet to be overcome.

It is easy to speculate why some of these inefficiencies still exist, and it is evident in the case of others. But no matter the reason, they collectively act as obstacles to getting it right the first time.

This is where specialists and trusted partners such as Kirby Group Engineering can significantly contribute to achieving this goal. We understand that the integrated design process is key to getting it right the first time.

An integrated design partnership of clients, engineers, constructors, and vendors provides the perfect recipe for the challenge.

“Get it right the first time” should be the warrior cry for data center construction leaders for the foreseeable future. But, back then, it was anything but.

Over the past ten years, I have witnessed a seismic shift in demand across the sector, ultimately driven by the global market for data, and only if we learn from the mistakes of our shared past can we be prepared to meet future demand. There are several actions we can take now that will provide the best chance at future proof solutions.

Firstly, we must address the increased need for more qualified, competent, and experienced human resources.

We are acutely aware of the importance of dealing with the lack of availability in the market. We have a multi-faceted strategy in place to deal with the issue head-on. It is a two-track approach to retention and recruitment.

Secondly, there is a requirement for far greater efficiencies in the project delivery process. As I have already alluded, we need a project delivery process to ‘get it right the first time’, that makes this possible.

We also need to maximize the exploitation of tools available to construct accurate digital twins of data center projects to ensure that inefficiencies are at least mitigated and hopefully eliminated. This is not possible, I believe, in the absence of an integrated design and project delivery process.

And lastly and probably most importantly, we need an investment in technology and technological-based project delivery tools.

The last ten years have shown a considerable lag in the correct and effective use of available technological-based project delivery tools across the industry.

For example, the availability of BIM 4D and 5D, while around for many years, is still not widely deployed. Reasons include a lack of available competent resources and lack of suitable graduate output at third level.

While we are seeing increased investment in this area, more needs to be done to attract prospective candidates into available programmes, while at the same time increasing capacity and flexible learning options within those programmes.

Kirby Group Engineering is already deploying training methodology with its investment in the Kirby Academy, which has dozens of online modules readily accessible.

It is committed to developing its people and invests an average of three days of learning and development into each employee every year. It has an uncompromising drive for continuous improvement and has developed the Kirby Way as a means by which process, and delivery are meticulously road mapped.

These initiatives are driven by Kirby’s core values of People, Quality, Safety, Delivery and Value – timeless concepts that will shape the future development of data center design and construction, regardless of any new innovations or technology.

It is also precisely these types of investments right across the industry that will reap the reward of lean production and on-time delivery over the next ten years.

In addition, we are in the process of introducing our in-house training programme for engineers on data center design and construction. We pride ourselves in the knowledge, expertise, and competency gained from operating very successfully in this industry.

We believe that now is the time to use this expertise at our disposal to produce world-class engineers that have been specifically trained and educated to meet our project and client needs and by way of continually improving our approach to investing in our people.

While I feel privileged to have played my part in the emergence of the data center sector in recent years, I am genuinely excited about what the future holds for me and my cohort of engineers working in this sector, and if we make incremental changes with best practices, we can effect real change.

Barry O'Sullivan, Group Electrical Engineering Manager at Kirby Group Engineering
Barry O'Sullivan, Group Electrical Engineering Manager at Kirby Group Engineering
Scroll to Top
Skip to content