Applying SRE Principles to CI/CD

Thursday 13:00

Continuous Integration and Deployment is the automation of building and testing code, CI/CD allows teams to ship code easily and frequently, with a high level of trust that end users won’t be impacted by bugs. What happens when painfully slow and unreliable CI/CD systems hinder our ability to deliver?

Site Reliability Engineers (SREs) guide their efforts with a set of agreed service level objectives (SLOs) and service level indicators (SLIs). A set of parameters that define what a system’s users should reasonably expect, and measure their performance. Join Buildkite’s Mel Kaulfuss to discover how to approach CI/CD with an SRE’s mindset. Learn what SLOs and SLIs are, and how to define them for your own build and deploy processes in order to create a shared understanding between a system’s stakeholders (most commonly its maintainers and users).

You’ll learn:

  • What SLOs and SLIs are
  • How to define SLOs and SLIs for your build, testing and deployment processes
  • How to use them to effectively drive continuous improvement
  • How to approach your CI/CD architecture just as you would your production systems:
    • Availability & Uptime on self managed cloud hosted servers
    • Performance & Speed with job parallelisation and autoscaling clusters
    • Abstraction & Extensibility by managing infrastructure as code
    • Observability & Insights via access to useful data, real time monitors and metrics

Mel Kaulfuss

Mel is a Developer Advocate at Buildkite. She has spent the past decade delivering software, either as Software Engineer, Production Coordinator or Project Manager. She loves fostering inclusive developer communities; has organised and emceed numerous RubyConfs in Australia, and most recently launched Buildkite’s own developer conference; UnblockConf. When she’s not clickity clacking, she’s patting dogs, eating strawberries, learning German and watching Nordic Noir.

GitHub Twitter