From Legacy Vendor Tooling To OTEL: Scaling Observability at MSCI With a Four-Person Team
Talkat Observability Day 2025 · on Tuesday, 2025-04-01· at 14:40· Level 1 | Hall Entrance S10 | Room C·
summary
Migrating a 5,000-person finance firm to an OpenTelemetry-centered stack that funnels logs, metrics, and traces into Prometheus, Jaeger, Grafana and Elasticsearch, implemented and operated by a four-person team, reduces mean time to detect by ~30% and sustains roughly 1 GB/s to Elasticsearch (about 2+ TB/day). A Kubernetes-deployed OTEL collector, automated onboarding scorecards and Helm-driven pipeline versioning yield 80% application coverage and lower cost with no headcount growth.
AI-written summary · about
chapters (14)
- 0:00MSCI and team intro
- 3:05Critical incident sparking change
- 5:09Legacy observability challenges
- 7:16OpenTelemetry as solution
- 8:59Why OpenTelemetry chosen
- 10:24Observability architecture design
- 13:15OpenTelemetry journey start
- 15:00Onboarding teams to OTel
- 16:25Scaling to one gigabyte per second
- 18:12Business impact and costs
- 19:58Organizational obstacles
- 21:03Future plans
- 22:48Key takeaways
- 25:15Q&A
related talks
37mHow To Adopt OpenTelemetry in an Enterprise Where Incumbent Vendor Tools Reign Supre...
34mPanel: OpenTelemetry: Bridging Platform and Enablement
36mPanel: OpenTelemetry...
49mAdding OpenTelemetry to Production Apps Lessons Learned
35mOpenTelemetry for Busy People
42mOpen Source for Better Observability
37mShould I Use OTel (collectors), or Is Prometheus Good Enough?
37mObservability For You And Me With Opentelemetry