Evolution of Observability Tools at Pinterest
Offered By: USENIX via YouTube
Course Description
Overview
Explore the evolution of observability tools at Pinterest in this 42-minute conference talk from SREcon19 Europe/Middle East/Africa. Discover how Pinterest's metrics system, log search, and distributed tracing adapted to meet changing requirements as the company scaled from a small startup to web-scale. Learn about the development of tools like Statsboard, Logsearch, and Pintrace, and gain insights into data reduction techniques, the TScript scripting language for time-series data, and integrated alerts. Examine the implementation of quick dashboards, automated canary analysis, and root cause analysis. Understand the automation roadmap, key lessons learned, and the role of the observability team at Pinterest. Gain valuable knowledge about scaling observability tools for large-scale web applications and the challenges faced during this evolution.
Syllabus
Intro
Pinterest
Change Cycle Change
Statsboard
Logsearch
Pintrace
Usage Growth
Tool Architecture
Data Reduction
TScript Scripting Language for Time-series
Integrated Alerts
Quick Dash
RunDash
Automated Canary Analysis
Automated Root Cause Analysis
Automation Roadmap
Lessons Learned
Observability Team
Acknowledgements
We're hiring! Come work with us!
Taught by
USENIX
Related Courses
How to Not Destroy Your Production Kubernetes ClustersUSENIX via YouTube SRE and ML - Why It Matters
USENIX via YouTube Knowledge and Power - A Sociotechnical Systems Discussion on the Future of SRE
USENIX via YouTube Tracing Bare Metal with OpenTelemetry
USENIX via YouTube Improving How We Observe Our Observability Data - Techniques for SREs
USENIX via YouTube