Resource attributes tell you who emitted the telemetry; they are how you slice signals by service, host, pod or version.
Resource attributes are attributes that describe the entity producing telemetry, such as service.name, service.version, host.name or k8s.pod.name, and are attached to all the signals that entity emits, so you can group and filter telemetry by its source.
Where span or metric attributes describe an individual operation, resource attributes describe the producer: which service, which version, which host, which Kubernetes pod and namespace. They are set once for the source and stamped onto every trace, metric and log it emits.
Resource attributes are how you attribute telemetry to its origin and correlate across signals from the same source. Filtering by service.name and service.version, for instance, lets you compare a new release against the old one, and enriching Kubernetes telemetry with pod and namespace attributes is what lets you pivot from a metric to the exact pod.
Sentinel AI uses resource attributes to attribute an incident to a specific service, version, host or pod, which is the first step in localising a fault.
Resource attributes describe the producer of telemetry (service, host, pod); span attributes describe an individual operation within a trace.
service.name, service.version, host.name and k8s.pod.name are common resource attributes attached to all of a source's telemetry.
Ops Singularity turns open telemetry into autonomous, governed resolution. See it on your own stack.