Technical presentation - 30 minutes (including q&a)
In this talk, we’ll explore a practical, repeatable process for tracking and evaluating Linux Kernel performance on ARM64-based systems in various environments. We will discuss the motivations behind continuous kernel performance evaluation, common challenges such as cloud environments and hardware diversity, and an end-to-end approach that integrates systematic methodologies, various past experiences from different teams and projects, automation and statistical rigor. Additionally, we’ll share our experience in building such a system atop a well-known open source and LKFT arsenal of tools to discover performance regressions before they become critical.