Gergely Szabó

IT Expert @ Origoss Solutions

Efficient System Monitoring in Cloud Native Environments

Prometheus (https://prometheus.io/) has become the standard solution for performance monitoring in cloud-native environments. It comes with a wide range of exporters that let you collect metrics starting from the hardware level up to the application level. Collecting and visualizing metrics exposed by the Linux kernel is a no-brainer with Prometheus. The recent kernel tracing developments allow even deeper insights into the Linux kernel in an efficient way. Technologies, like eBPF and the BPF Compiler Collection (bcc), have emerged to help out the initiated few: Linux kernel developers and performance engineers. The presentation will walk you through the basic concepts of eBPF. You will learn how far you can get with it. You will see how much effort it demands to get started. Yet eBPF can be a valuable tool for us 'mere mortals'. We will demonstrate how to enrich Prometheus with deep kernel metrics to get better observability of your cloud system.

Gergely Szabó is an IT expert at Origoss Solutions. He spent more than 15 years in the IT industry as researcher, developer, architect, system engineer, etc. In the last few years he has been focusing on IT cloud, more specifically on the cloud-native technologies, like Cloud Foundry, Kubernetes and OpenShift.

Gergely is Certified Kubernetes Administrator and Application Developer. His philosophy is centered around observability which is important in the cloud-native world as never before.