Operations is ready for change

Rick Cronin, Technical Director, EMEA, Skytap, discusses key takeaways from the 2016 DevOps Focus Groups roundtable event.

Skytap recently had the privilege of sponsoring a series of DevOps Focus Groups in London where we had the opportunity to host three roundtable discussions with numerous executives from the corporate IT sector.

Events like this are a great way to get a pulse reading on what’s driving – and stalling – the modernisation journeys that so many large enterprise organisations are undertaking.

The topic of discussion that we posed to each group was “DevOps and the cloud,” with a focus on trying to gauge if companies are more likely to embrace modernising their infrastructure and then their processes and culture, or vice versa. By the end of the day, while most of whom we had the pleasure of meeting had not yet actually begun organisational shifts to DevOps or the cloud, many were on their way, and their collective reasoning for wanting to make the transformation was both spot‑on and inspiring.

Overcoming resistance to the cloud

Our roundtables’ executives came from a variety of industries – financial services, travel, telecommunications, insurance, education, etc., and each were at various early, evaluative stages of ‘application modernisation.’ We learnt that many large enterprises are looking at a variety of approaches to updating core IT systems, and we discovered that using the cloud to help modernise traditional applications is a popular approach.

However, there’s often much resistance within large enterprises to attempt to modernise these types of monolithic applications, and with good reason. These applications are often the backbone of an enterprise, and are responsible for core business functions such as the systems of record that cannot be disrupted.

“We know we need to modernise for tomorrow, but we also have to keep the business running today,” said one of our focus group’s attendees. “We really can’t be too disruptive, because we’re also responsible for safe‑guarding revenue.”

This concern is not only valid – nearly everyone throughout the day had similar apprehensions – it’s also not only limited to the UK. Our view, and one that we’ve seen prove very effective, is a structured, iterative, three‑step approach that reduces risk, and increases innovation.

Skytap often recommends these best practices:

  • Removing delivery bottlenecks caused by resource scarcity.
  • Introducing modern processes and supporting tooling.
  • Employing modern cloud technology and architectural patterns.

This approach allows development, testing, and operations teams to more easily begin and complete infrastructure, process and architecture modernisation journey initiatives.

Keeping pace with born-in-the-cloud applications

Depending on who you ask – especially if it’s developers and testers – IT operations teams have been much more associated with being bottlenecks rather than removing them, but that way of thinking is rapidly changing.

Today’s forward‑thinking operations teams, like those who attended the DevOps Focus Groups, are looking for ways to increase test coverage, decrease time to market, and modernise legacy systems in order to keep pace with newer, born‑in‑the‑cloud applications.

“We have to bring modern technology to our older apps, because if we don’t modernise our infrastructure today, we won’t be around tomorrow,” said one attendee.

Real world examples of DevOps demand

We asked our groups why they were looking at DevOps and the cloud to deliver a collaborative culture, and modernise their infrastructure, architecture, and applications as a whole and many said it was because the speed of business demands it. There were some fantastic, real world examples of this demand, including:

“I have to be able to respond to requirement demands from customers and internal resources. It currently takes us too long to maintain and operate our systems.”

“We need the ability to do software releases without taking platforms offline. There are some releases where we have to bring the website down for four hours. It’s crazy that we’re in a world where we have to do that.” „

“From an operations point of view, it’s really about having the ability to just say, ‘Yes’ when we’re asked for a new process, tooling, infrastructure, anything that we’re asked for. Today we’re like, ‘Not right now, no,’ because we don’t have the flexibility of infrastructure. Everything’s just static.”

The need for DevOps champions

Providing that level of flexibility is a monumental task, but it was so inspiring to meet so many executives who clearly see the incredible opportunity before them to drive not just continued stability, but disruption that results in real business change. We asked each of our focus groups, “Who are the DevOps champions in your organisations? Who is driving adoption?” and the answers were not surprising.

While the groups’ thoughts were echoed throughout the day, one attendee summarised it by saying, “I think it’s the IT people on the ground. The people with the transformation of thinking, ‘I am an IT champion,’ to, ‘I am a business champion.’ is really important. I do everything for business. IT can’t function alone. That kind of thought process is very important.”

There have been a million definitions of DevOps, but I’d like to suggest that “DevOps is about creating and sustaining teams that take software all the way from ideas, to customer use, to end of life. These teams absolutely change in size and makeup over time, but it’s important that it feels like one collective team is fully responsible for – and empowered to deliver – a successful offering.”


Cultures, technologies, and processes like DevOps, cloud, continuous integration and delivery, etc., are massively transformative changes for many organisations to rally around, but they offer the potential for returns on investment that can be seen very early on. Whether the drumbeat to adopt these shifts comes from the ground‑up, or top‑down, from what we’ve seen, IT has the chance to lead the way and not just meet business demands, but also consistently exceed them.


This article originally appeared in the DevOps Focus Groups Supplement as part of the January 2017 issue of TEST Magazine. Edited for web by Jordan Platt.