Quick Navigation: A | C | E | F | G | H | I | L | M | O | P | R | S | T | U | W

A#

Alerting Rules#

Rules in Prometheus that define condition-based alerts. When conditions are met, alerts are sent to Alertmanager.

Alertmanager#

A component that receives alerts from Prometheus and groups, inhibits, and routes them.

C#

Cardinality#

The number of unique time series. Higher cardinality occurs with more label combinations.

Context Propagation#

A mechanism for passing Trace ID and Span ID between services in distributed systems.

Counter#

A monotonically increasing metric type. Used for request counts, error counts, etc. Calculate rate of change with rate() function.

E#

Exemplar#

A trace sample linked to a metric. Allows direct navigation from metrics to related traces.

Exporter#

A component that exposes application/system metrics in Prometheus format.

F#

Four Golden Signals#

Four core metrics proposed by Google SRE: Latency, Traffic, Errors, Saturation.

G#

Gauge#

A metric type representing current value. Can increase/decrease. Used for CPU usage, temperature, etc.

H#

Histogram#

A metric type that measures value distribution in buckets. Used for measuring response time distribution. Calculate percentiles with histogram_quantile().

I#

Instrumentation#

Adding observability data collection code to applications. Can be automatic or manual instrumentation.

irate()#

A PromQL function that calculates instantaneous rate using only the last two samples.

L#

Label#

Key-value metadata attached to metrics. Used for filtering and grouping.

LogQL#

Grafana Loki’s query language. Similar syntax to PromQL.

Loki#

Grafana’s log collection system. Lightweight with label-based indexing.

M#

Micrometer#

A metrics facade for JVM applications. Supports various backends like Prometheus, Datadog.

O#

OpenTelemetry (OTel)#

A vendor-neutral observability standard framework for metrics, logs, and traces.

OTLP#

OpenTelemetry Protocol. A standard protocol for transmitting observability data.

P#

Percentile#

A value below which a certain percentage of data falls in the distribution. P99 = 99% of values are below this.

PromQL#

Prometheus Query Language. A language for querying and analyzing time series data.

Pull Model#

A method where Prometheus actively scrapes targets for metrics (opposite of Push).

R#

rate()#

A PromQL function that calculates the average per-second rate of increase for Counters.

Recording Rules#

Prometheus rules that pre-calculate complex queries and store them as new metrics.

RED Method#

A microservice monitoring methodology measuring Rate, Errors, Duration.

S#

Sampling#

A technique that stores only a portion of all traces. Used for cost optimization.

Scrape#

The act of Prometheus collecting metrics from targets.

Service Level Indicator (SLI)#

A metric that measures service level. Example: P99 response time, error rate.

Service Level Objective (SLO)#

Target values for SLIs. Example: P99 < 500ms, 99.9% availability.

Span#

A single unit of work in distributed tracing. A Trace consists of multiple Spans.

T#

Tail-based Sampling#

A sampling method that prioritizes storing errors/slow requests after request completion.

Tempo#

Grafana’s distributed tracing backend. Optimized for large-scale trace storage.

Three Pillars#

The three pillars of Observability: Metrics, Logs, Traces.

Trace#

The complete path of a single request through a distributed system. Consists of multiple Spans.

Trace ID#

A unique ID identifying a trace. All Spans share the same Trace ID.

U#

USE Method#

A resource monitoring methodology measuring Utilization, Saturation, Errors.

W#

W3C Trace Context#

An HTTP header standard for distributed tracing. Uses traceparent header.