Shared agentsLogs
All indexed, no sampling.
- Observability & Monitoring
//Traces
Distributed traces, end to end.Every span retained.
OpenTelemetry-native, no sampling. End-to-end visibility into your distributed systems.
Correlation
Correlated telemetry
Metrics, logs, and traces troubleshot in a single pane of glass.
Storage
Every span, kept
Apache Parquet compression keeps every span without losing query speed.
OpenTelemetry
OTel-native traces
Instrument once with the standard SDK, with zero vendor lock-in.
//Overview
What is distributed tracing?
Distributed tracing follows a single request as it moves through services, queues and databases, recording each step as a span, so teams can see where time is spent and where errors start.
DataObserve Sense is OpenTelemetry-native: it keeps every span without sampling, maps service dependencies automatically and links traces to the logs and metrics around them, from the browser to the database.
//Follow a request wherever it goes
From collection to the exact span
Auto-instrumentation
Zero-code instrumentation configured by the collector.
Dependency mapping
Real-time service maps reveal how your microservices truly connect.
Performance insights
Latency heatmaps and error trends, compared across versions.
Detailed spans
Precise timing, tags, and logs, one click from overview metrics.
Context propagation
W3C trace context, with flame graphs and waterfall diagrams.
Flexible retention
Object storage as your hot tier, no complex tiering.
//Explore more withDataObserve
More from Sense · Observability & Monitoring
Metrics
PromQL and SQL, full resolution, retained.
- Observability & Monitoring
Frontend Monitoring
Real user monitoring, session to stack trace.
- Observability & Monitoring
Synthetic Monitoring
API and browser checks, global or private, unified with the rest of your observability.
- Observability & Monitoring
//FAQs
Traces: questions teams ask
Is it OpenTelemetry-native?
Yes. OTLP over HTTP or gRPC from any SDK or collector, with no proprietary agent.
Will I be locked in?
No. Your instrumentation never changes if you switch backends later; only the exporter endpoint does.
Do frontend errors link to backend traces?
Yes. RUM and backend tracing share trace context, so a frontend error links to the span that caused it.
What is a span?
A span is one timed operation within a trace, such as an HTTP call or database query. A trace is the full tree of spans for one request.
Does tracing require code changes?
Not always. OpenTelemetry auto-instrumentation covers many languages and frameworks without code changes; custom spans can be added where you need more detail.
Why keep every span instead of sampling?
Sampling can drop the rare slow or failing request you need to debug. Keeping every span means the evidence is always there.
