Tracing Transaction Latency in Cloud-Native Systems with OpenTelemetry

In cloud-native microservices, a single user request often spans multiple services, databases, and external dependencies. Small delays across these components accumulate, making latency issues difficult to diagnose using logs or metrics alone.
This session focuses on how to trace and understand transaction latency across distributed systems using OpenTelemetry.
We will explore how request-level tracing provides visibility into where time is actually spent, and how it helps engineers identify bottlenecks that are not visible through traditional monitoring approaches.
Topics covered:
• What transaction latency means in microservices systems
• Why logs and metrics are insufficient for latency debugging
• Fundamentals of distributed tracing
• Using OpenTelemetry to trace requests across services
• Identifying bottlenecks using span-level visibility
• Translating trace insights into performance improvements
This session is designed for engineers working with:
• Kubernetes-based platforms
• microservices architectures
• distributed systems
• observability and performance engineering
If you are debugging latency issues or trying to understand how requests behave across services in production, this session will provide practical techniques and a repeatable approach.