To Find Everyone and Bring Them Together: Event Tracing System in VK Calls

on YouTube ↗41m

" summary

VK Calls implemented an event tracing system to diagnose call issues like failed connections, dropouts, and missing screen sharing. The system collects and links call events and traces across components to make root-cause investigation practical instead of manual, ad-hoc searching.

AI-written summary · about

" chapters (13)

1. 0:00Introduction
2. 1:50VK Calls user support issues
3. 7:06Tracing system reasons
4. 9:15Palandir requirements
5. 10:40Palandir architecture
6. 14:55Palandir interface
7. 17:30Data types and events
8. 18:57Data volume and load
9. 20:01Palandir implementation results
10. 21:03Case analysis before/after
11. 27:18System development plans
12. 30:28Summary and advantages
13. 31:53Q&A

" related talks

• 54mEverything Everywhere at Once: Tracing under High Load with 10k BackendsAleksandr Kirsanov · Saint HighLoad++ 2023
• 01h 30mIntroducing CallTracing(tm), based on RabbitMQ, Spring and ZipkinMonish Unni · SpringOne2GX 2015
• 39mTracing VMs Seeing How Host and Guest InteractSteven Rostedt · LinuxCon Europe 2022
• 44mTracer — an Alternative to Crashlytics. A Different Take on App Error AnalyticsKirill Popov · Mobius 2023 Autumn
• 57mDistributed Tracing: How the Pros Debug Concurrent and Distributed SystemsAaron Stannard · NDC London 2019
• 53mFrom Algorithm to Production: How to Approach Verification of Distributed SystemsNikita Galushko · Saint HighLoad++ 2023
• 01h 00mBuilding your own profiling and diagnosis tools with Event Tracing for WindowsPavel Yosifovich · DotNext 2019 Piter
• 07mWhy Tracing is a Lifesaver for E2E Tests of Distributed Systems?Burak Kantarci · TestJS Summit 2021