Developer Voices Building Observable Systems with eBPF and Linux (with Mohammed Aboullaite)
16 snips
Oct 31, 2025 Mohammed Aboullaite, a backend engineer at Spotify, dives into the world of continuous profiling and observability using eBPF. He explains how eBPF allows for monitoring at the kernel level without recompiling, making it a game-changer for managing large-scale systems. Listeners learn about the evolution of eBPF, how to handle data volumes from profiling, and practical tools like Pyroscope and Pixie. Mohammed also discusses the low overhead of eBPF and the importance of mapping kernel events back to code, providing invaluable insights for optimizing performance in production.
AI Snips
Chapters
Transcript
Episode notes
Camping Next To The Server Rack
- Kris recounts camping by a server rack as a joke to illustrate old-school monitoring during the dotcom boom.
- The story highlights how monitoring expectations have risen since those ad-hoc days.
Kernel-Level Observability With eBPF
- eBPF lets you extend the Linux kernel by loading verified bytecode programs at runtime without recompiling the kernel.
- That kernel-level visibility unlocks language-agnostic observability across processes and resources.
Shrink Fidelity Over Time
- Reduce stored profiling fidelity over time: keep high-frequency recent data and coarser historical snapshots.
- This tiered approach cuts storage cost while preserving actionable recent detail.