Live Site Culture & Site Reliability at Azure DevOps

on YouTube ↗01h 05m

" summary

Azure DevOps describes a live‑site system that collects 7 terabytes per day of telemetry, surfaces it via an in‑app diagnostics bar, and routes alerts to a small distributed SRE team plus rotating feature live site engineers (two per feature team) who own mitigation and backlog fixes. Automated mitigations include an auto mitigator and widespread circuit‑breaker patterns; status updates target 15‑minute distribution and a $600 paid escalation filters urgent customer wakeups.

AI-written summary · about

" chapters (10)

1. 0:00Start
2. 6:03Knight Capital
3. 13:11Azure DevOps Live Site culture
4. 18:17Our Live Site process
5. 27:10Transparency builds trust
6. 29:43Driving down time to communicate
7. 32:52Our telemetry pipeline
8. 35:09Instrumenting the Customer Experience
9. 36:59Service Hygiene - error (and perf) budgets.
10. 38:47Extend visibility to dependencies

" related talks

• 01h 02mEffective DevOps for OrganizationsAnthony Borton · NDC Sydney 2022
• 54mUnraveling the DifferencesChris Ayers · PowerShell + DevOps Global Summit 2024
• 21mBuilding A DevOps Culture in a Remote WorldEmily Freeman · DevOpsDays Chicago 2020
• 01h 02mBuild a DevOps Culture: Microsoft's Journey to adopt an Agile Mindset and DevOps cultureMickey Gousset · NDC Oslo 2020
• 46mAutomated Testing Using Azure DevOpsRoman Nedzelsky · TestCon Europe 2021
• 46mReal World DevOpsAbel Wang · Build Stuff 2020
• 44mDevOps at Microsoft: Our Internal Transformation StoryApril Edwards · IDI Incontro DevOps Italia 2020
• 01h 03mA lap around Azure DevopsJakob Ehn · NDC London 2019