System Design Space
Knowledge graphSettings

Updated: February 21, 2026 at 11:59 PM

eBPF: The Documentary

hard

Unlocking The Kernel - how Linux kernel extension technology changed networking, security and observability.

eBPF: The Documentary

Unlocking The Kernel - the history of the creation of revolutionary technology

Year:2023
Production:Isovalent

Source

Book cube

Film review from Alexander Polomodov

Перейти на сайт

What is eBPF

eBPF (extended Berkeley Packet Filter) is a technology for safely and efficiently extending the capabilities of the operating system at runtime. Initially it was only Linux, but then the technology was repeated for other operating systems.

The point is that to expand the capabilities of the kernel no changes to the kernel source code required, and there is no need to load individual modules. Security is achieved through a verifier inside the kernel, which performs static code analysis.

Loaded programs that pass the test are either interpreted or JIT-compiled for performance. The execution model is event driven—programs are attached to various hook points in the kernel and run when an event occurs.

Related topic

Site Reliability Engineering

Reliability and observability practices from Google

Читать обзор

History of creation

💡 Idea

It all started with Alexei Starovoitov, while working at the startup PLUMgrid on SDN (Software Defined Network), he came up with the idea of embedding a virtual machine into the Linux kernel. He created a new instruction set based on x86 assembly with security verification in the kernel.

🔄 Evolutionary path

The idea was too revolutionary and Chris Wright from Red Hat (now CTO of Red Hat) proposed an evolutionary approach - to integrate new technology into the existing BPF subsystem. This is where the name "extended Berkeley Packet Filter" comes from.

📊 Tracing

Initially the focus was on networks, but there was already a solution for this problem. Then the focus shifted to tracing - Brendan Gregg created the tools that became bcc and bpftrace.

🚀 Hyperscalers

eBPF has shown effectiveness in Meta, Google, Netflix and other hyperscalers. Alexei Starovoitov joined Facebook in 2015 and showed outstanding results there.

Ecosystem Development

🐝 Cilium

Isovalent created Cilium, a project that brought eBPF capabilities to Kubernetes end users.

🔒 Security

eBPF has found application in security - system call monitoring, anomaly detection and runtime protection.

📈 Observability

eBPF-based tools allow you to deeply profile systems without significantly impacting performance.

🌐 Cross-platform

The technology was repeated for other operating systems - eBPF went beyond Linux.

Key people

Alexei Starovoitov

Creator of eBPF. Worked at PLUMgrid, then at Facebook/Meta.

Brendan Gregg

Productivity Expert. Creator of bcc and bpftrace.

Chris Wright

CTO Red Hat. He proposed an evolutionary path of integration into the core.

Impact on the industry

We managed to implement the original idea and change the state of affairs in the industry. Previously, writing code to run inside the kernel (rather than in userspace) was extremely difficult, time-consuming, and expensive. Now this can be done quickly and easily using eBPF.

Enable tracking in Settings

System Design Space

© 2026 Alexander Polomodov