SRE: An incomplete guide to cultural Narnia
summary
Company-scale SRE at Bloomberg handles 4,000 developers and thousands of apps by splitting roles into application SREs and infrastructure SREs, building centralized services (HBase Hadoop Mesos), and enforcing telemetry and log aggregation and run books. Application SREs spend roughly a 70/30 split between team work and platform work, backed by a tiered frontline support that fields pages and improves playbooks so core engineers avoid noise and infrastructure teams avoid midnight surprises.
AI-written summary · about
chapters (20)
- 0:00Intro
- 0:40About me
- 1:14SRE
- 1:59Disclaimer
- 2:42Fun terms
- 3:08Its not about technology
- 4:17Cultural shift
- 4:55What is SRE
- 5:41Areas of responsibility
- 6:24Application SRE
- 7:54Bridging the gap
- 9:20Where did we get
- 10:12Application
- 11:45Infrastructure
- 12:51Organizational Layout
- 13:59SRE Example
- 15:02Support Structure
- 16:12Frontline Support
- 19:41Conclusion
- 20:09Contact info
related talks
25mSRE in the Third Age
26mFrom Nothing to SRE: Practical Guidance on Implementing SRE in Smaller Organisations
22mBuilding SRE: Culture from the Outside In
30mDifferences in SRE Implementations across Companies
43mSRE From Inception to Global Scale
25mSRE Guiding Principles & Transformational Framework
35mCare and Feeding of SRE
40mSRE as She Is Spoke