The missing piece in your observability puzzle!
summary
Continuous profiling fills the observability gap by running low-overhead profilers in production (claimed 2–5% overhead) and shipping time-series profiles for long-term analysis. It uses language profilers like Java Flight Recorder and async-profiler plus kernel-level collection via eBPF-based collection, addresses JVM safe‑point bias, and faces big challenges in symbolization and storage. Open-source options include Parca, Periscope, Pixie; a Periscope demo produces flame graphs that pinpoint a mutex hotspot in EU‑North and validates the approach end to end.
AI-written summary · about
chapters (17)
- 0:00Introduction and speaker intro
- 1:50Evolution of application architectures
- 4:44Challenges in distributed systems observability
- 7:57Three pillars of observability
- 13:48Limitations of traditional observability
- 18:06Traditional profiling challenges
- 21:37Why production is the happiest place
- 23:05What is continuous profiling
- 27:03Challenges in continuous profiling
- 30:37Profiling techniques and types
- 34:05Benefits of continuous profiling
- 37:39Acquiring profiles in applications
- 42:02Introduction to eBPF
- 44:34Open source continuous profiling tools
- 48:05Future of continuous profiling
- 49:11Demo: Flame graphs in Periscope
- 52:08Q&A and closing remarks
related talks
56mContinuous Profiling the missing piece in your observability puzzle
40mContinuous Profiling, the missing piece in your observability puzzle!
01h 13mContinuous Profiling: The Missing Piece in Your Observability Puzzle
46mContinuous Profiling - the 4th Pillar of Observability
24mCommunity Driven Breakout Session: Continuous Profiling
36mContinuous Profiling in the Cloud-Native era
36mProfiling in the cloud-native era - Matthias Loibl
36mContinuous Profiling with Parca- Beneath the Kernel to Beyond...