How we at Avito analyze 5 million traces and conduct architectural oversight
summary
Avito describes processing five million distributed traces to build a service-interaction map of their microservices using tracing tools and related infrastructure. They explain the tooling and methods used to analyze that interaction map to detect architectural problems and improve system reliability.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 3:21Architectural oversight problem
- 11:27First trace analysis prototype
- 13:15NeFJ performance issues
- 15:49NeFJ query optimization
- 17:14Cyclic dependency detection
- 18:39Synchronous loop requests
- 21:04Long call chains
- 22:50Service criticality assessment
- 29:35Summary and next steps
- 32:36Q&A
related talks
51mHow We Do Tracing at the Scale of Thousands of Services and Millions of Spans per Second
52mMonitoring for Developers! Technology for the Community! Profit for Everyone!
32mA Picture is Worth 1,000 Traces
45mMonitoring in a Microservice Architecture
29mHow we tricked engineers into utilizing distributed tracing
21mNavigating the Complexities of Service to Service Invocations: Deep and Brief Dive Into Causality
51mNavigating Distributed Traces
37mThrough the looking glass: Effective observability for cloud native applications
