logo

Essential Patterns For Designing And Implementing Your Operator

2022-10-26

Authors:   Michael Hrivnak, Austin Macdonald


Summary

Lessons learned in designing and implementing operators for Kubernetes
  • API anti-patterns to avoid in operator design
  • Challenges in bridging the gap between imperative infrastructure management and declarative Kubernetes API
  • Maximizing usefulness of client's cache and avoiding memory bloat
  • Efficiently interacting with multiple clusters from a single operator instance
  • Minimizing load imposed on the API server
The speaker shared an anecdote about a common anti-pattern of putting important information in annotations instead of the spec of a resource, which can lead to future regret.

Abstract

It’s easy to get started developing operators with kubebuilder and operator-sdk to manage your workloads and infrastructure – but what challenges will you face as your operator matures? This presentation will share the most essential lessons learned across years of experience helping teams and organizations design and implement real-world operators for a wide variety of use cases. Coding topics will focus on Go-based operators. You will learn about: - API anti-patterns: Common API design choices that lead to future regret, and how to overcome them in the wild. - Bridging the gap between slow (and complex and buggy) imperative infrastructure management and the declarative Kubernetes API. - Taking control of the client’s cache to maximize its usefulness and avoid memory bloat. - Interacting with multiple clusters efficiently from a single operator instance. - Minimizing load imposed on the API server. Attendees will be ready to face key challenges as they enhance their operators with new features and evolving APIs.

Materials:

Post a comment

Related work

Authors: Jonathan Berkhahn, Per Gonçalves da Silva
2023-04-21


Authors: Rashmi Gottipati, Varsha Prasad Narsing, Tony J, Bryce Palmer, Camila Macedo
2023-04-19