Scaling SLOs with Kubernetes and Cloud Native Observability (EN)
Defining Service Level Objectives and Service Level Indicators is a really important aspect of implementing SRE. Through service metrics (SLOs, SLIs, Error Budgets), SRE can help us measure our system’s performance and improve customer experience. They not only enable your teams to monitor and plan around reliability, but can also be early predictors of customer satisfaction, NPS, churn rates, and more. With the rise of cloud native technologies, it has become more and more relevant to automate our observability, extending it to an SLO-as-code model. In this session we’ll see how SLOs have evolved and can be used in a Cloud Native world. We’ll then explore how technologies like Kubernetes and Prometheus can help us scale SLOs, while promoting best practices and standards using Observability as code. Finally, we’ll see how to put all these together with Jenkins and Rancher, to operationalize error budgets.
George HantzarasCitrix Systems
George is a distributed systems expert and a hands-on engineering leader with focus on delivering B2B cloud services at scale. He is a Director of Cloud Platform Engineering at Citrix, coming from a DevOps background, he focuses on implementing SRE at enterprise scale working with cloud native technologies. He has been organizing the Athens Cloud Computing Meetup since 2016. Most recently, he has been a speaker at global events like Hashiconf, DeveloperWeek, Voxxed Days, DevNexus and more.