DevOpsCulture

As technologists we’ve spent decades getting good at developing and delivering efficiencies, reducing labor costs, and implementing automation. So it’s no surprise that the part of DevOps we do best is automating our own tools. However, we often drop the ball on the parts of the pipeline that lie outside the software - the “hard” decisions that humans need to make around licensing, billing, scaling costs, etc.

One of the principles of Agile, DevOps, and Extreme Programming is to favor continuous integration and continuous deployment. To avoid monolithic pileups that are disruptive both the the software as well as the team, we should be practicing small, fast, Lean releases, ideally through Continuous Delivery. Once way to measure this is to check release size. We can measure the size of the release as a ratio of the number of lines changed in a release over the the total lines of code in the project. The smaller the amount of change in the release, then the more frequently we...

Breaking down silos and sharing knowledge across a team improves the understanding of the codebase by the entire team, and accelerates the ability for the team to react to changes or outages. Having bottlenecks where only one team member understands a portion of the code results in risky scenarios where the team is held up from being able to delivery quickly. This is the Hit By A Bus Factor. How many of your team members are indespensible to the project? That is, if they were hit by a bus, would work come to a stop?

An agile team is receptive to change. And a good indication of this is that the team is improving the quality of their code over time. This should manifest through continuous refactoring. We can objectively measure code refactoring by looking for code change where unit tests do not require modification, and the unit tests continue to pass. That indicates that the code

Due to the rapid pace of market change and customer demand, Agility is the single most important indicator of performance in a modern organization. As large organizations adopt digital transform strategies, and as small technology companies grow, it becomes imperitive to find some way to measure the agility of teams in a consistent objective way.