datarekha
MLOps Medium Asked at GoogleAsked at DatadogAsked at AmazonAsked at Uber

What is the difference between monitoring and distributed tracing?

For MLOps Engineerbackend-engineerAI / LLM Engineer
The short answer

Monitoring aggregates health signals such as request rate, error rate and latency percentiles across systems and time, detecting that a population is unhealthy. Distributed tracing follows one request through nested spans across services, localizing where it waited or failed. Metrics trigger investigation; traces explain individual paths.

How to think about it

Monitoring uses fleet-level metrics, dashboards and alerts: p95 latency, error rate, saturation, queue age and cost. It answers is something wrong, and how broad is it?

Tracing records one request as related spans across gateways, services, databases and model calls. It answers where did this request spend time or fail?

Learn it properly System design boundaries

Keep practising

What is the difference between leading and lagging indicators, and how do you use both when building a metrics system? How do you monitor a model when ground-truth labels are delayed or never arrive? A model is live and you cannot get labels quickly. How do you set up alerting to catch performance problems early? What's the difference between experiment tracking and a model registry, and why do you need both? DAU dropped 15 % week-over-week with no planned changes. How do you diagnose it?
All MLOps questions

Explore further

Observability & tracing OpenTelemetry for GenAI Inference metrics: TTFT, ITL & goodput
OLAP (Online Analytical Processing) Streaming Kafka Sharding
Skip to content