eBPF’s Abilities and Limitations: The Truth - Liz Rice & John Fastabend, Isovalent

Don't miss out! Join us at our next Flagship Conference: KubeCon + CloudNativeCon North America in Salt Lake City from November 12 - 15, 2024. Connect with our current graduated, incubating, and sandbox projects as the community gathers to further the education and advancement of cloud native computing. Learn more at kubecon.io
eBPF’s Abilities and Limitations: The Truth - Liz Rice & John Fastabend, Isovalent
eBPF is proving to be a great platform for cloud native infrastructure tooling, with several CNCF projects leveraging it to implement networking, security and observability capabilities from within the kernel. But as with any new technology, there are various myths and uncertainties circulating about it in the community, particularly around its limitations: you might hear that it’s not Turing complete, that it can’t be used for anything that involves state, or that it can’t be used to parse Layer 7 protocols. In this talk we’ll disprove all these rumors with demonstrations including: - Looping in eBPF - Leveraging maps for state - An eBPF implementation of a Turing machine equivalent This doesn’t mean eBPF is the right hammer for every nail; using the Cilium project as an example we’ll discuss why not every feature is implemented in the kernel. (Yet?)