Book Image

Accelerate DevOps with GitHub

By : Michael Kaufmann
Book Image

Accelerate DevOps with GitHub

By: Michael Kaufmann

Overview of this book

This practical guide to DevOps uses GitHub as the DevOps platform and shows how you can leverage the power of GitHub for collaboration, lean management, and secure and fast software delivery. The chapters provide simple solutions to common problems, thereby helping teams that are already on their DevOps journey to further advance into DevOps and speed up their software delivery performance. From finding the right metrics to measure your success to learning from other teams’ success stories without merely copying what they’ve done, this book has it all in one place. As you advance, you’ll find out how you can leverage the power of GitHub to accelerate your value delivery – by making work visible with GitHub Projects, measuring the right metrics with GitHub Insights, using solid and proven engineering practices with GitHub Actions and Advanced Security, and moving to event-based and loosely coupled software architecture. By the end of this GitHub book, you'll have understood what factors influence software delivery performance and how you can measure your capabilities, thus realizing where you stand in your journey and how you can move forward.
Table of Contents (31 chapters)
1
Part 1: Lean Management and Collaboration
7
Part 2: Engineering DevOps Practices
14
Part 3: Release with Confidence
19
Part 4: Software Architecture
22
Part 5: Lean Product Management
25
Part 6: GitHub for your Enterprise

Picking the right migration strategy

When migrating to a new platform, you have different options:

  • High-fidelity migration: You try to migrate as much as possible to the new platform.
  • Clean cut-over migration: You only migrate the bare minimum that is necessary to start working on the new platform.

High-fidelity migrations to complex platforms have different problems. The main problem is that there is not a 1-to-1 mapping of all entities and that things just work differently on different platforms. By migrating everything over, you influence the way people use the new system. The data is optimized for the old system using old processes. Also, the time, costs, and complexity involved in a high-fidelity migration are not linear. The more you try to get to 100% fidelity, the more complex and expensive it gets, and 100% is normally not achievable at all (see Figure 21.1).

Figure 21.1 – Complexity, time, and costs for different levels of fidelity...