The cloud-first strategies of businesses across sectors are maturing, with DevOps teams playing a leading role More than half (51%) of financial services businesses in Europe have a strategy to put IT in the cloud wherever possible, and 84% expected to do so within two years, with DevOps teams leading the way.

According to a survey of 102 senior executives by ReRez Research for cloud software supplier ServiceNow, this puts financial services above the average for cloud adoption across all business sectors, which is 77%.

The survey, of 1,850 executives in the US, the UK, Australia, Singapore, Germany, France and the Netherlands, covered the healthcare, manufacturing, professional services and retail sectors, as well as financial services.

Despite the charge to the cloud, there are challenges ahead for companies in all sectors when moving to the cloud, particularly around staffing. Some 89% of companies in all sectors that have completed the shift to a cloud-first model said their current IT staff lacked the required skillsets to be successful in the new cloud world.

DevOps is driving cloud take-up in financial services, according to 67% of the finance firms questioned in the survey.

“For years, pundits have waited for an enterprise shift from traditional datacentre computing to cloud computing,” said Chris Pope, office of the chief strategy officer at ServiceNow. “Today, reality meets hype in the financial services sector, and cloud-first consumption will accelerate at breakneck speed in the next two years.

“Cloud is not new, so why was 2016 the year that enterprises became cloud-first? According to the survey, the answer is the rise of DevOps. DevOps saw early that cloud computing could solve many of the issues they were grappling with. And DevOps’ ‘can-do’ attitude – enabled by cloud – is extending across the enterprise.”

Meanwhile, CIOs that are already using the cloud are preparing for a hybrid future. The 2017 Computer Weekly/TechTarget IT Priorities poll revealed that CIOs are preparing the hybrid-readiness of their datacentre facilities.

The research found that connecting on-premise datacentre assets to public cloud resources is a top investment priority for UK and European IT decision makers in 2017.

A Brief Look at DevOps

DevOps grew out of the Agile movement, which applied streamlined, iterative practices to software development. DevOps extended those ideas into IT operations, combining development and operations into a single, unified workflow. Its defining technical characteristics include:

• Elimination of silos — development, testing, security, and operations teams collaborate rather than work in isolation.
• Infrastructure as Code (IaC) — infrastructure is defined, versioned, and managed through code rather than manual configuration.
• Automation — provisioning, testing, and deployment are scripted and repeatable.
• Continuous delivery — software moves through the pipeline in a steady, automated flow rather than in large, infrequent releases.
• Virtualisation and containers — applications are deployed in virtual machines or lightweight containers rather than directly on physical hardware.

A significant current emphasis in DevOps practice is the shift toward full containerisation — moving away from monolithic applications and toward microservices running in containers such as Docker, orchestrated by platforms such as Kubernetes.

A Brief Look at Cloud Computing

The cloud emerged from advances in processor speed, memory, storage, and network bandwidth that made it practical to pool large numbers of servers into a virtualised, on-demand resource pool. Major technology vendors had attempted various forms of utility computing for years; the cloud finally made the model economically and technically viable at scale.

As cloud adoption grew, it naturally gave rise to new ways of developing and operating software. Teams that once managed physical racks found themselves working entirely through APIs and dashboards. Infrastructure became something you rented by the hour and configured through code — not something you bolted together and manually maintained.

Whether DevOps shaped cloud practices or cloud computing shaped DevOps is a question for technology historians. What is clear is that the two have co-evolved to the point where many DevOps practices are simply descriptions of how cloud environments already work.

Virtualisation: Cloud Makes It the Default

The cloud is a virtual environment by definition. Cloud users — including the engineers deploying and operating applications — never interact directly with bare metal. Several layers of abstraction sit between the user and the underlying hardware.

In that context, DevOps’s emphasis on virtualisation stops being a recommendation and becomes a simple description of reality. Deploying applications in virtual machines or containers inside a cloud environment is just another layer of abstraction on top of an already-virtualised platform. The performance overhead, when managed well, is minimal.

The move to containerisation is particularly well-suited to the cloud. Containers are lightweight, portable, and start quickly — qualities that directly complement the elastic, on-demand nature of cloud infrastructure. Running containers on cloud virtual machines gives teams fine-grained control over resource usage while keeping deployment fast and reproducible.

Infrastructure as Code: Cloud Makes It Unavoidable

One of the most consequential DevOps principles is treating infrastructure as code: defining servers, networks, storage, and configuration in version-controlled files rather than through manual, point-and-click processes. The arguments for doing so are strong:

• Manual infrastructure management is slow, error-prone, and expensive.
• It encourages the creation of snowflake systems — one-off, specially configured machines that are difficult to reproduce, audit, or replace.
• Scripted, automated infrastructure is consistent, auditable, and repeatable.

On physical hardware, achieving full infrastructure automation can be a significant undertaking. In the cloud, it is simply how things work. Cloud providers expose their entire platform through APIs. Tools such as Terraform, AWS CloudFormation, Azure Bicep, and Google Cloud Deployment Manager let engineers define entire environments in code. There is no physical server to rack, no cable to plug in, and no manual step that cannot be scripted.

In a cloud environment, infrastructure as code is not an aspirational practice — it is the path of least resistance.

Continuous Delivery: Cloud Removes the Logistical Barriers

Continuous delivery — the practice of keeping software in a permanently releasable state and deploying updates frequently — was always a logical goal for internet-delivered software. The historical barrier was logistical: testing and distributing software for diverse desktop environments, or coordinating releases across physical data centres, introduced delays that made frequent releases impractical.

Cloud deployment removes most of those barriers. Deploying an update can be as simple as running a script that pushes a new container image or updates a cloud function. There is no physical media to ship, no heterogeneous desktop environments to support, and no manual server patching to coordinate. The deployment target is a single, well-defined, API-accessible environment.

That simplicity makes continuous delivery not just possible but natural. The pipeline from code commit to production becomes a fully automated, scriptable workflow with no inherent delays.

Parallel Testing: A Cloud-Enabled Acceleration

In traditional software development, testing is often the biggest source of delay. Sequential test runs on a fixed set of machines can add days or weeks to a release cycle. The cloud changes the economics of testing fundamentally.

Because cloud virtual machines can be provisioned on demand and released when no longer needed, test suites that once had to run sequentially can now run in parallel across many instances simultaneously. The practical effects are significant:

• Test runs that previously took days can complete in hours.
• Tests that previously took weeks may take a day or less.
• Test coverage that was previously omitted due to time constraints can be included without a meaningful impact on overall pipeline duration.

This acceleration directly enables continuous delivery. Faster, broader testing means teams can ship more frequently without sacrificing quality — the combination that makes continuous delivery a genuine improvement rather than a shortcut.

The Sum of the Parts: Why DevOps and Cloud Belong Together

Taken individually, each of the connections above is meaningful. Together, they describe a relationship that goes beyond compatibility. In a cloud environment:

• Virtualisation is not a DevOps recommendation — it is the environment you are working in.
• Infrastructure as code is not an aspirational practice — it is the primary way cloud resources are managed.
• Continuous delivery is not a difficult cultural shift — it is the logical result of removing deployment friction.
• Parallel testing is not a luxury — it is an affordable default enabled by on-demand compute.

DevOps principles that can feel like idealistic advocacy in an on-premises context turn out to be straightforward descriptions of how cloud platforms already operate. That alignment is why organisations adopting cloud infrastructure almost inevitably find themselves adopting DevOps practices alongside it — and why those that adopt DevOps practices find cloud infrastructure a far more natural operating environment than traditional data centres.

If DevOps had not grown from the Agile movement, something functionally equivalent would almost certainly have emerged from cloud deployment practices on its own. The match is that close.