Taming Observability at Scale in a Multi-Cluster Kubernetes Platform at Bloom... Joe Nathan Abellard
Kubernetes Karmada Prometheus Thanos Observability Multi Cluster kubernetes Multi Region architecture Sre Site reliability engineering Cloud native Kubernetes platform Metrics aggregation
This talk presents Bloomberg's enterprise-grade approach to observability in a large-scale, multi-cluster Kubernetes platform built on Karmada. It covers practical solutions for aggregating metrics across multiple Prometheus servers using Thanos, strategies for reducing alert fatigue while maintaining timely issue detection, and real-world challenges of defining SLIs and SLOs in a multi-tenant environment. Ideal for platform engineers, SREs, and DevOps professionals managing production Kubernetes clusters at scale.
Related videos
Lightning Talk: Beyond Billions: Operating Thanos, Prometheus & OpenTelemetry... Narendra Sanikommu
Keynote: Cloud Native at Bharat Scale, How Rapido Scaled to 7+ Mi... Srivatsa Katta & Adarsh K Kumar
The Full Picture: Visualizing Service "Fullness" To Rethink Saturation Prevention - Tal Nordan
When the Cloud Fails: Debugging the "Undocumented" - Dhruv Jain, Gojek (GoTo Group) Indonesia
AI-Powered Root Cause Analysis at Scale: From Theory To Production... Letícia Mota & Yevgeny Gladun
Planet-scale Dashboards at Google with Grafana
k8s-monitoring-helm Chart Office Hours (April 2026)
Rab Ne Bana Di Jodi - Why Platforms & AI Need Each Other - Ram Iyengar, Cloud Foundry Foundation
20 minutes to 2 minutes: How Home Depot automated incident responses with OpenTelemetry
From Running to Operating: Real-World Strimzi in Production | Rajith Attapattu, Randoli
Grafana Customer Testimonial - Philip Pencal Alter Domus
