D2DO252: (Re)Building Cloudflare’s Millions-of-Logs-Per-Second Logging Pipeline

Colin Douch, Observability Tech Lead at Cloudflare, and Jayson Cena, SRE at Cloudflare, dive into the complexities of migrating from Syslog-NG to OpenTelemetry. They discuss the motivations for this shift, such as scalability and memory safety, while tackling challenges like maintaining uninterrupted customer traffic. The duo also highlights the importance of redundancy in logging systems and shares insights on logging protocols, illustrating the balance between resource usage and operational speed in a high-performance environment.