DevOps Guidance - DevOps Guidance
IntroductionDefinitions

DevOps Guidance

Publication date: September 20, 2023 (Document history)

Drawing from Amazon's own transformative journey and the expertise gained by AWS in managing cloud services at global scale, the AWS Well-Architected Framework DevOps Guidance offers a structured approach that organizations of all sizes can follow to cultivate a high-velocity, security-focused culture capable of delivering substantial business value using modern technologies and DevOps best practices.

Introduction

In the early 2000s, Amazon went through its own DevOps transformation which led to an online bookstore forming the Amazon Web Services (AWS) cloud computing division. Today, AWS provides a wide range of products and services for global customers that are powered by that same innovative DevOps approach. Due to the positive effects of this transformation, AWS recognizes the significance of DevOps and has been at the forefront of its adoption and implementation.

Amazon's own journey, along with the collective experience gained from assisting customers as they modernize and migrate to the cloud, provided insight into the capabilities which we believe make DevOps adoption successful. With these learnings, we created a collection of modern capabilities that together form a comprehensive approach to designing, developing, securing, and efficiently operating software at cloud scale.

The content and recommendations outlined within the Well-Architected DevOps Guidance are based on our expertise operating and managing services for global customers and serves as a customizable reference to fit your organization's requirements. Each organization has unique requirements which may deviate from the examples provided here.

There is no one-size-fits-all approach to adopting DevOps. Tailor these recommendations to suit your individual environment, quality, and security needs.

Definitions

  • DevOps Sagas are core domains within the software delivery process that collectively form AWS DevOps best practices. Together, they form a collection of modern capabilities representing a comprehensive approach to designing, developing, securing, and efficiently operating software at cloud scale. You can use the DevOps Sagas as a common definition of what DevOps means to align on a shared understanding within your organization.

  • Capabilities are repeatable mechanisms that an organization can continuously improve to sustainably practice DevOps and achieve measurable outcomes. Each capability includes best practice indicators, metrics, and anti-patterns that can be tracked and refined over time. Capabilities encompass a wide range of topics related to the associated DevOps Saga. They are not presented in any specific order.

  • Indicators are a collective term covering best practices spanning people, process, and technology. They can be used as qualitative measurements, such as a checklist, to objectively assess and continuously track your organization's ability to perform a capability. Each indicator contains a prescriptive improvement plan that contains high-level guidance for performing it. Indicators can fall into one of three categories, which represent importance:

    • Foundational: Requirements to achieve minimum viability of the capability when practicing DevOps. Indicators in this category are considered essential for organizations to meet to achieve basic proficiency of the capability.

    • Recommended: Recommended best practices and advanced techniques that can improve the performance of the capability. Indicators in this category offer significant benefits, such as improved security, developer experience, speed, reliability, or the ability to optimize implementation in cloud-based environments.

    • Optional: Minor adjustments and improvements that further enhance the ability to implement the capability or apply only to specific use cases. Indicators in this category might not be applicable to all organizations or workloads, but can provide additional benefits to some.

  • Anti-patterns are practices that might seem beneficial initially, but have the potential to lead to unexpected or negative outcomes. They are indicators that we have seen slow down or halt our customer's progress when adopting DevOps within their organizations. Identifying and addressing anti-patterns can help you to avoid pitfalls and optimize the chance for successful DevOps adoption.

  • Metrics are quantifiable measures of an organization's ability to perform the associated capability. Think of indicators as the how and metrics as the how well when using this guidance to measure your organization's ability to implement a capability. There is no one-size-fits-all approach to selecting which metrics to track. We provide metrics for each capability as a starting point. Each organization needs to determine which metrics are important to their business. Consider using additional frameworks, such as DevOps Research and Assessment (DORA) and SPACE, to further customize the metrics we provide so that they align to your organizational goals.

For more Well-Architected terminology, see Definitions in the AWS Well-Architected Framework whitepaper.