Reference
Glossary
The vocabulary of application monitoring, one term per page, each defined in a sentence first.
- APM (application performance monitoring)
- The practice and the tooling for measuring how a running application behaves for its users, chiefly response time, error rate and throughput, and tracing slow or failed work to its cause.
- Availability test
- A synthetic check that Azure Monitor runs against a URL from several Azure locations on a schedule, recording each run in the availabilityResults table.
- Cloud role name and instance
- The two fields that say which service (cloud_RoleName) and which machine, container or instance (cloud_RoleInstance) sent a telemetry item; Application Map draws one node per role name.
- Connection string
- The setting that tells an Application Insights SDK or OpenTelemetry distro where to send telemetry, as semicolon-separated key=value pairs holding the instrumentation key and the regional endpoints.
- Daily cap
- A limit on the GB a Log Analytics workspace or Application Insights resource ingests in a day; once it is reached, collection stops until the next daily reset.
- Dependency
- A call an app makes to something outside it, such as an HTTP API, a database, a queue or a cache, recorded in the dependencies table with its target, duration and result.
- Exporter
- The OpenTelemetry component that sends finished spans, log records and metrics out of an app to a backend, such as Azure Monitor, an OTLP endpoint or the console.
- Instrumentation key
- The GUID that identifies an Application Insights resource; it used to be the whole configuration and is now one part of the connection string.
- itemCount
- The column on every sampled Application Insights row that records how many original telemetry items the row stands for, so totals stay right after sampling.
- Live Metrics
- Application Insights' real-time view of incoming requests, failures, dependency calls and server performance, streamed from the app and never stored.
- Log Analytics workspace
- The Azure Monitor data store behind workspace-based Application Insights; telemetry lands in its App* tables, and billing, retention and the daily cap are set there.
- OpenTelemetry Collector
- A standalone OpenTelemetry service that receives telemetry, runs it through pipelines of processors and exports it to one or more backends.
- operation_Id
- The Application Insights column that ties together every row of one distributed operation; in OpenTelemetry and W3C Trace Context terms it is the trace ID.
- OTLP (OpenTelemetry Protocol)
- OpenTelemetry's protocol for sending traces, metrics and logs over gRPC or HTTP, which Azure Monitor now accepts natively through the OpenTelemetry Collector.
- Percentile (p50, p95, p99)
- The value a given share of measurements fall at or below; a p95 latency of 800 ms means 95% of requests took 800 ms or less, and one in twenty took longer.
- Sampling
- Keeping a representative fraction of telemetry instead of all of it, to cut ingestion volume and cost while keeping counts and percentiles statistically useful.
- Standard and log-based metrics
- Application Insights' two kinds of built-in metric; standard metrics are pre-aggregated as telemetry is collected and unaffected by sampling, log-based metrics are queries over stored rows that sampling does affect.
- Telemetry channel
- The classic Application Insights SDK component that buffers telemetry items and sends them to the ingestion endpoint in batches; OpenTelemetry calls the same job an exporter.
- Telemetry initializer
- A hook in the classic Application Insights SDK that runs on every telemetry item before it is sent, used to add or correct properties such as the cloud role name, user ID or a custom dimension.
- Telemetry processor
- A link in the classic Application Insights SDK's processing chain that sees each telemetry item before it is sent and can change it, pass it on or drop it; sampling is built as one.