DevOps: The pillars of operational excellence

DevOps is the practise of combining the development work and operations work of a software project. Its goal is to shorten the development lifecycle and is often thought of as the realm of just the weird tech geeks. Or at best, that it only affects the technical team – making deployments safer and less time-consuming.

This is a myth. When we believe that DevOps only relates to the technical teams, we underestimate the value of DevOps and we will never experience its true strength and power. This is where Amazon Web Services’ (AWS’s) well-architected pillar of operational excellence has such a critical role to play in bringing DevOps to the attention of the entire organisation.

AWS defines operational excellence as a combination of areas:

• The organisation
• Preparation
• Operation
• Evolution

While each of these areas could have entire essays written about them, they all boil down to the same three concepts that the DevOps Handbook encourages:

1. The principles of flow
2. The principles of feedback
3. The principles of continuous learning

In order to fully explain and understand these sets of principles, entire books could be written. In fact, the “Phoenix Project” and the “DevOps Handbook” are highly recommended reading for anyone beginning their DevOps journey. However, for the sake of clarity, the “three ways” can be explained as:

• Enabling fast left-to-right flow of work from Dev through Ops and, finally, to the customer. This is the area that most people think DevOps end as it includes all the famous practices of Continuous Integration and Continuous Deployment (the infamous CICD pipeline).
• Enabling a fast and constant flow of feedback and information from right to left. In other words, we aim to create a fast and efficient feedback loop in the software delivery pipeline. This enables teams to detect problems quicker and resolve them more efficiently. This helps to create more resilient and safer systems for both customers and developers.
• Creating a high-trust culture and team dynamic where mistakes are corrected without blame and distributing these learnings across the entire company. This assists in creating an environment where cumulative experience and learning has a multiplied effect on success and quality – creating a high-pace, high morale team.

Just a brief glance at these principles is enough to see how DevOps is not the responsibility of a single person or team (or even a single department such as Dev or Ops). Rather, it’s a cross-domain and cross-functional initiative that brings in every single area and needs to be implemented across the entire company.

Just a brief glance at these principles is enough to see how DevOps is not the responsibility of a single person or team.

As many blogs, presentations and talks have said: “If you have a DevOps team… you’re doing DevOps wrong.” The goal of DevOps is not to automate the Ops teams out of a job. It is to create a fundamentally different way of delivering value to our customers.

And so, we return to the initial question…. How can AWS’s well-architected pillar of operational excellence help in creating a DevOps transformation across an organisation? I feel the answer should be clearer now. Operational excellence is aimed at CTOs and higher executives.

While the implementation of the practices and policies rely on the whole enterprise, the strategy of operational excellence is in the hands of much higher-up executives. In other words, every area of the company has a stake in achieving operational excellence in some way. When we look at the areas highlighted by AWS’s own white paper, we can start to see how they line up with the three ways of DevOps:

Looks like DevOps and operational excellence line up very nicely? While it’s not entirely true to say that DevOps is the same as operational excellence (or that DevOps is not possible in a non-AWS environment), the tools and services that AWS recommends when achieving operational excellence make DevOps a much easier accomplishment.

From Config, SecurityHub, Trusted Advisor and Control Tower to CloudWatch, CloudTrail and CloudFormation (not even referencing the Developer tools); the entire DevOps lifecycle is taken care of. Additionally, these tools are aimed at every team member in the organisation − from managers to developers, operations, security and everything in between.

And so – who is responsible for operational excellence? The answer seems obvious now. As with DevOps, operational excellence cannot be the domain of a single person or team. It is the responsibility of the entire organisation.

As these two concepts aim to kick-start innovation across the organisation, the entire organisation needs to contribute and be a part of this incredible and inspiring journey.

Clearly, this brief overview skims through a large amount of detail and a DevOps transformation is a journey that will not be completed in a short space of time. However, as we move into more and more uncharted and faster changing operating environments, it is something that every organisation should undertake as soon as possible.