“When done correctly, an operating model introduces new ways to provide higher levels of agility and operational excellence. It really is the essence of how work gets done.”

“Operating models aren’t static. They’re dynamic, based on the needs and requirements that exist today, as well as the nature of competitive forces, external factors and variables that may be out of your control.”

“Understand the business capabilities of I&O and how you define value. That ultimately leads to a higher degree of differentiation and is the first step towards a much more refined IT operating model.”

“If I&O were to remain partitioned off as just a technical element of value, it will become disjointed from the business strategy over time.”

“While org charts are familiar and simple to understand, they usually don’t reflect how work gets done or how decisions are made.”

“Start at a service level as it has immediate value to the business. It enables engagement with other lines of business and allows you to be viewed as more than an I&O organization that provides infrastructure.”

Value Stream Platforms: The end-to-end sequence of activities performed to deliver value to the customer, from building and deploying an application to maintaining it in production. There are 2 types: 1) value stream management platforms and 2) value stream delivery platforms.

AI augmented devops: The use of AI-enabled technologies to augment the entire software development life cycle. One example of this is email platforms auto generated or auto complete a user’s draft sentences. 

Continuous infrastructure automation: This trend applies agile and DevOps practices such as version control, continuous testing and CI/CD to infrastructure code rather than application code, bringing the same benefits to apps teams and I&O teams alike. 

Gitops: A technique for operating a cloud native application using only declarative constructs stored in Git, which is increasingly popular with organizations that are using Kubernetes and containers.  

Site Reliability Engineering (SRE): A software engineering mindset that uses automation to reduce toil, or low value tasks, as the guiding principle.

Chaos engineering: The use of experimental and potentially destructive failure testing in pre-production to uncover vulnerabilities and weaknesses (e.g., single points of failure, latency) within a complex production system. 

“The fleeting, on-demand nature of modern environments creates great challenges for IT operations in understanding how the applications and services over the top of these actually perform. We have to radically redesign how we think about monitoring these systems.” 

“To get insight into how your applications are performing, focus on how end-users are interacting with them. Understand their journey through the system, rather than the individual components of it, and how their experience ties into the underlying environment.”

“What makes containers great, makes them very hard to manage. A very large number of containers have a very short lifespan of minutes, sometimes seconds. This completely changes how you have to think about monitoring.”

“Don’t monitor something simply because it exists. Do it in a way that is meaningful and relevant to your end-users, critical applications and services in your environment.”

“Investigate open source and commercial observability tools. Use tools to help you align where your needs are versus what can be provided by each of these tools. Do this with an eye to the future.”

“Skill up your team. This is as much about the people as it is about the tools. Look to broaden skill sets to move beyond the traditional IT operations mindset of just working in an operations environment, to something that is much more agile and DevOps aligned.”