Community Driven Breakout Session: Continuous Profiling
summary
Continuous profiling captures hot paths, memory and CPU over time and correlates with metrics and traces using tools like pprof endpoint and pyroscope, while kernel-level sampling via eBPF promises language-agnostic visibility. Implementation pain points include Ruby thread‑local context sharing and a donated Datadog profiler that lacks current maintainers. Practical wins include finding a serverless workload where profiling revealed roughly 50% CPU spent in service mesh. Standardizing on pprof-compatible output would let more runtimes interoperate and speed adoption.
AI-written summary · about
chapters (9)
- 0:00Introduction
- 2:05What is Continuous Profiling
- 4:22How do you use Continuous Profiling
- 5:12Challenges with Continuous Profiling
- 9:13Service Mesh
- 11:33Aggregation
- 17:08Is PPROF useful
- 19:51Datadog donation
- 21:50hackathons
related talks
01h 13mContinuous Profiling: The Missing Piece in Your Observability Puzzle
56mContinuous Profiling the missing piece in your observability puzzle
40mContinuous Profiling, the missing piece in your observability puzzle!
52mThe missing piece in your observability puzzle!
36mContinuous Profiling in the Cloud-Native era
26mProfiling Mastery: Decode Your Data and Elevate Observability
24mWhy Does Continuous Profiling Matter to Developers?
41mGain a deeper understanding of your incidents: continuous profiling