Devops for Growth

This quasi-continuous release cycle comes with several advantages:

It eliminates the need for hotfixes. In the three-push-a-day system, if a critical change had to get out and it wasn’t during one of the scheduled push times, someone had to call for a hotfix. These out-of-band pushes were disruptive because they usually needed some human action and could bump into the next scheduled push. With the new system, the vast majority of things that would have required a hotfix can simply be committed to master and pushed in the next release.

It allows better support for a global engineering team. We tried to schedule the three daily pushes to accommodate our engineering offices around the world, but even with that effort the weekly push required all engineers to pay attention at a specific date and time that was not always convenient in their time zone. The new quasi-continuous system means all engineers everywhere in the world can develop and deliver their code when it makes sense for them.

It provides a forcing function to develop the next generation of tools, automation, and processes necessary to allow the company to scale. When we take on projects like this, it works as a pressure test across many teams and systems. We made improvements to our push tools, our diff review tools, our testing infrastructure, our capacity management system, our traffic routing systems, and many other areas. These teams all came together because they wanted to see the main project of a faster push cycle succeed. The improvements we made will help ensure the company is ready for future growth.

It makes the user experience better, faster. When it takes days or weeks to see how code will behave, engineers may have already moved on to something new. With continuous delivery, engineers don’t have to wait a week or longer to get feedback about a change they made. They can learn more quickly what doesn’t work, and deliver small enhancements as soon as they are ready instead of waiting for the next big release. From an infrastructure perspective, this new system puts us in a much better position to react to rare events that might impact people. Ultimately, this brings engineers closer to users and improves both product development and product reliability.

Ce plan de livraison sera mis à jour au cours de chaque revue de sprint (« Sprint Review ») en tenant compte de la mise à jour de l’indicateur. Ceci permet, à la fin de chaque sprint, d’inspecter et d’adapter (« Inspect & Adapt ») le carnet de produit et de réorienter les efforts.