If you asked a child today what a cloud is, what response do you think you’d get? “Water drops floating in the air”? Probably, yes. However, some would opt for a different response: “Servers accessed over the internet”.  Interestingly, if you google ‘What is a cloud’ the first responses are not the meteorological definition. Modern cloud computing, which in absolute timespan is still very young, has become a pretty much ubiquitous facet in our lives. Most of us now store a majority of our personal data on numerable cloud servers operated by huge companies who are now household brands. Any remaining confusion in nomenclature and etymology doesn’t reside with the young. For those of us old enough to remember floppy disks, we can still get confused by terminology. Hopefully none of us believe that cloud storage implies harnessing previously unknown capacities of the humble H2O…

Tongue out of cheek, we should certainly forgive those who get befuddled by the jargon. Like many concepts in technology, cloud computing has many overlapping definitions and it can be difficult to contain. Microsoft offer an elegant one “Simply put, cloud computing is the delivery of computing services – including servers, storage, databases, networking, software, analytics and intelligence – over the Internet (“the cloud”) to offer faster innovation, flexible resources and economies of scale”. Whilst that is an accurate precis of modern cloud computing, its roots are quite different. Like much of our technology, the roots of cloud computing were planted out of a military expedient. Building on John McCarthy’s work on time-sharing in the 1950s, in 1966, J.C.R Licklider et al initiated the ARPANET project. In the height of the cold war, the U.S. Defence Department was keen to find a way to de-centralise command and control infrastructure. ARPANET intended to solve this problem by ‘packet switching’, a method for transmitting data between computers in discrete locations connected by phone lines. ARPANET was not only the father of modern cloud computing, but the foundational technology behind the internet itself.

If we skip through a few generations, mainly shaped by the ongoing creation of the world-wide web, we get to the mid-1990s. Some seek to credit Google or Amazon with birthing the ‘cloud’ definition in the mid noughties, but we prefer to look to George Favaloro and Sean O’Sullivan at Compaq in 1996. Their key foresight was that enterprise software would eventually cede to web-enabled services. Prescient indeed. A few years later, a pioneering Oracle salesperson decided to take a leap inspired by a vision. Few could dispute the impact that Salesforce and Marc Benioff have personally had on the development of cloud computing. Inspired by a hatred of floppy disks and clumsy local installations, their “End of Software Campaign” in 2000 was the inception of Software-as-a-Service. Just look at what SaaS has grown up to be today

The backstory and provenance of cloud computing is vital to understand as it shapes the way enterprise architecture is implemented today and defines aspirations for the future. It has not been long since monolithic, on-premise enterprise architecture was the predominant way that organisations developed their IT infrastructure. Cloud computing has been integral in creating new paradigms. SaaS, PaaS, IaaS, Microservices and DevOps are interconnected but distinct means of delivering the fundamental benefits achieved via developing distributed computing systems. All describe ways that organisations are now leaning into more nimble, agile and ultimately more productive ways to build.

So why hasn’t the adoption of cloud computing been universal across the enterprise landscape? Still today, many organisations persist traditional on-premise builds. For example, some large banks, insurers and accountancy firms we work with remain reticent to make the switch. Information security has and continues to be a material concern for enterprise sites that process large quantities of sensitive client data. Negative consequences associated to info-sec breaches, both from a regulatory and reputational perspective, are so material that they have prevented these firms from making the switch. Another connected but distinct concern is loss of infrastructure control linked to switching to the cloud. Again, for organisations who are audited with respect to their IT operations, porting BAU workflow to the cloud has proven to be problematic. Having actual ‘brick and mortar’ walls around servers that can be serviced by in-house engineers is still considered mandatory for some.

However, even within these heavily regulated and necessarily conservative firms, we are witnessing real change. Over the past few years, we have worked with a UK insurer who have made it a strategic priority to move their entire systems architecture off-premise and into the cloud. The motivating factors for this company are a microcosm: DevOps is critical to them as it supports continuous incremental improvement which is critical to their business model. Microservices are knitting together a new ‘best-of-breed’ architecture to replace a monstrous monolithic legacy. Elastic scaling is vital to process larger analytical requests as they acquire lumpy new books of assets. The ongoing servicing and maintenance of a cloud-based infrastructure is optimal as they seek to sustain a strategic rather than solely operationally focussed IT department.

This example crystallises and brings to the forefront the concept of ‘cloud nativity’. Indeed, what our pioneering Insurance client is doing is laying the foundations for a cutting-edge cloud native infrastructure. What’s the difference between cloud and cloud native? In our opinion, you won’t find a better, more succinct explanation than this one from Joe Beda at VMware. As Joe explains embracing managed services, be they at the infrastructure or at higher levels of abstraction is one thing. To truly consider oneself cloud native, one must embrace all the utility ‘native’ to cloud-based technology.

Before we consider the benefits, we should start with the challenges associated to a cloud native approach – as they are not trivial. To embrace a cloud native build you need to consider three key pillars: Containers, orchestration and microservices. Starting in reverse order, microservices are already the predominant delivery mechanism for modern software applications and are shaping the way that modern enterprise sites build in a best-of-breed, DevOps manner. Containers and cloud native orchestration are much more challenging concepts for corporate IT teams to integrate into legacy setups. Even if your physical infrastructure is no longer on-premise, re-factoring core code into Docker containers and replacing hypervisors with Kubernetes is burdensome to put it mildly.

So is it worth the hassle? It absolutely is. Enterprise IT stacks are riddled with complexity, but investment made in the direction of cloud nativity will reap dividends. In the post Covid era, which will be shaped by ongoing tidal waves of digital transformation, incumbent firms cannot afford to miss out on what cloud native offers. The ability to build new infrastructure quickly in an agile, customer-focussed and organisationally aligned way is vital to future commercial success. Client facing applications built in a stateless, non-server sticky way lead to big gains in customer experience. Containerisation abstracts away many of the emerging issues and costs latent in developing a multi-cloud, OS agnostic strategy. Finally and critically, cloud native as an approach mitigates vendor lock-in, which we know has been a significant operational pain-point for the firms we have worked with in the past few years.

We’ve come a long way in a short time with modern cloud computing. We encourage our clients and all financial markets enterprise sights to reach for the stars and embed cloud native aspirations at the heart of their tech strategies. Concerns remain germane; we spoke recently to a global investment bank who have until recently still insisted on local installations – but the winds of change are blowing. Perhaps the single biggest challenge for cloud native advocates is to continue to lobby for the organisational change required for implementation. For those organisations that accede to building a modern, cloud super-powered infrastructure, the future should sparkle indeed.