System Comprehension and Root Cause Analysis With Distributed Tracing

Event: Observability Practitioners Summit @ KubeCon/CloudNativeCon NA 2018.

Date: December 10, 2018

Speakers: Yuri Shkuro and Joe Farro

Video: YouTube (or below)

Slides: PDF

In this talk we discuss a data mining and visualization technique that allows Uber to gain operational insights and assist on-call engineers in root cause analysis by analyzing billions of traces we collect, not just a handful that power users of tracing happened to review.