Glossary
Quick Lookup
Common terms used in this module:
- System Monitoring - observing system behavior using operational metrics
- Signal - a raw or derived metric used to understand system behavior
- Anomaly - an observation or condition that deviates from expectations
- System State - an overall assessment of system condition
- Operational Decision - an action informed by monitoring evidence
System Monitoring
The practice of observing system behavior using measurements. Monitoring systems help detect issues early and maintain reliability. Examples:
- request volume
- error rates
- response times
Signal
A signal is a raw or derived metric used to understand system behavior.
Signals may be direct measurements or values calculated from other metrics. Well-designed signals help analysts recognize patterns, detect anomalies, and assess system performance.
Examples include:
- request volume
- error rate
- average latency
- throughput
Anomaly
An anomaly is an observation or condition that deviates from expected system behavior.
Anomalies depend on context, domain knowledge, and defined expectations.
An anomaly does not necessarily indicate a failure. It identifies a condition that may warrant investigation.
Examples include:
- unusually high error rates
- unusually slow response times
- unexpected drops in requests
System State
The system state summarizes the overall condition of a monitored system.
System states translate monitoring signals into interpretable assessments.
State classifications depend on the monitoring rules and operational context.
Example classifications include:
- STABLE - system behavior is within defined expectations
- DEGRADED - system performance or reliability is reduced
- FAILURE - the system cannot provide one or more required functions
An overall stable state does not necessarily mean that every individual observation is free of anomalies.
Operational Decision
An action taken in response to monitoring information. Examples include:
- investigating system errors
- scaling infrastructure
- rolling back a deployment
Continuous intelligence systems support operational decisions by providing timely information about system behavior.