Fueling innovative software
July 15-18, 2019
Portland, OR

Observability and performance analysis with BPF

David Calavera (Netlify)
1:45pm2:25pm Wednesday, July 17, 2019
Emerging Languages and Frameworks
Location: Portland 256
Secondary topics:  Open Source
Average rating: ***..
(3.20, 5 ratings)

Who is this presentation for?

  • Systems engineers, kernel engineers, and performance engineers

Level

Advanced

Description

BPF is a virtual machine inside the Linux kernel that provides secure and high performant observability with minimal overhead. It allows engineers to modify the kernel’s behaviors to certain events without having to build kernel modules or having to recompile the kernel itself. BPF is changing how engineers analyze and observe programs running at scale in production.

David Calavera demystifies BPF, which can be very intimidating because it’s a technology built within the Linux kernel, by showing examples to encourage you to learn BPF. He explores why this is a technology you want to have in your toolbox for observability and performance analysis, how BPF integrates with cloud native projects like Prometheus and Kubernetes to give you deep visibility inside your production infrastructure, and what types of programs you can write with BPF. You’ll leave prepared to dive into the topic yourself without being intimidated.

Prerequisite knowledge

  • A basic understanding of C or another modern programming language, such as Go
  • General knowledge of Linux kernel internals (useful but not required)

What you'll learn

  • Learn that the Linux kernel can be approachable and the Linux kernel's toolbox can be very useful for performance analysis
  • Understand that by using BPF engineers can gather knowledge about their applications in a new and very significant way
Photo of David Calavera

David Calavera

Netlify

David Calavera is the CTO of Netlify, where he and his team are building the best platform for deploying and automating modern web projects. Previously, he was a core member of the Docker Engine project, where he helped developers build the container engine that started the container revolution. David also built enterprise tools for GitHub and has contributed to numerous open source projects such us Go, JRuby, and many others.