How companies use LOKI to run their business
summary
Proximus runs a large Loki stack: 15 OpenStack clusters each with roughly 900 computes, 3 PB storage, 50,000 vCPUs and ~4,000 VMs, and 10 OpenShift clusters on OpenStack hosting ~400 worker nodes, 10,000 pods and 40,000 containers across 89 apps (≈800 microservices). Virtualized firewalls, Octavia load balancers, PCI passthrough, SR-IOV and CPU pinning handle telco NFV needs, Ixia CloudLens provides virtual tapping, infra-as-code and cross-functional teams enable yearly site-isolation upgrades and support bursty workloads of hundreds of thousands of connections.
AI-written summary · about
chapters (4)
related talks
15mOperations report: Running Loki on Different On-Prem Environments
31mLoki: Like Prometheus But for Logs
25mLoki: Prometheus-inspired logging for cloud native
29m5 Years in, Loki Turns 3.0
31mRunning the Infra at FOSDEM
05mGrafana Loki Log Aggregation for Incident Investigations
41mMonitoring OVH: 300k Servers, 27 DCs, a Metrics Platform
28mFrom Fragmented Logs to Unified Insights: Scaling O11y