threshold
A chosen limit that a value is compared against to decide whether something should happen, such as an alert firing.
In more detail
Thresholds appear in alerting rules, in log levels (records below the threshold are dropped) and in latency SLIs (requests under a threshold count as good). A threshold is useful only when someone chose and owns it.
Example
Threshold rules count events in a sliding window.Related terms
- alarm — A rule that watches a metric and changes state, and notifies someone, when the metric crosses a threshold for long enough.
- AWS Budgets — An AWS billing feature that compares actual or forecast spend with an amount you set and alerts people when it crosses a threshold.
- percentile — The value below which a given percentage of measurements fall; p99 latency is the time 99 in 100 requests beat.
- alerting rule — A Prometheus rule with a PromQL expression that describes an unhealthy condition, plus a for duration, labels and annotations.
- log level — A label on each log message saying how serious it is, from debug and info up to error, critical and emergency.
- sliding window — The most recent k items of a sequence or stream, which moves forward one item at a time as new data arrives.
- correlation rule — A detection rule that links several events, often from different sources, that together suggest an attack.
- low and slow attack — An attack spread out over time, with few attempts per hour, so it never reaches any count-based alert limit.
- anomaly detection — Learning what normal behaviour looks like for a workload or identity and flagging what is new.
- alert — An automatic notification sent when a monitored condition, such as a high error rate, stays true.
- autoscaling — A cloud feature that automatically adds servers when demand rises and removes them when it falls.
Taught in
- Prompts As Code — AI Engineering: LLM Apps, RAG & Agents
- Memory And Compaction — AI Engineering: LLM Apps, RAG & Agents
- What An Embedding Is — AI Engineering: LLM Apps, RAG & Agents
- Embedding Models — AI Engineering: LLM Apps, RAG & Agents
- Measuring Retrieval — AI Engineering: LLM Apps, RAG & Agents
- Workflows Before Agents — AI Engineering: LLM Apps, RAG & Agents
- Excessive Agency — AI Engineering: LLM Apps, RAG & Agents
- Eval-Driven Development — AI Engineering: LLM Apps, RAG & Agents
- Model-Graded Evals — AI Engineering: LLM Apps, RAG & Agents
- Evaluating RAG And Agents — AI Engineering: LLM Apps, RAG & Agents
- Regression Tests In CI — AI Engineering: LLM Apps, RAG & Agents
- Tracing And Monitoring — AI Engineering: LLM Apps, RAG & Agents
- Images, Audio And Speech — AI Engineering: LLM Apps, RAG & Agents
- Two Hundred Accounts And One Bad Deploy — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Strategies Across Compute: The Canary That Saw Nothing — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Finding The Problem: The Composite Alarm That Stayed Quiet — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Monitoring As Code: The Alarm That Watched Nothing — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Security Hub, GuardDuty And Inspector: Signal, Noise And Response — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Pipeline Security: The Build Role That Could Do Anything — AWS Certified DevOps Engineer – Professional (DOP-C02) Exam Prep
- Observability with CloudWatch — Cloud Infrastructure & Hyperscaler Architecture
- Metrics, Logs, and Alarms — Cloud Infrastructure & Hyperscaler Architecture
- Cost, Budgets, and Tagging — Cloud Infrastructure & Hyperscaler Architecture
- The Three Signals: Logs, Metrics And Traces — Full-Stack Observability: Metrics, Tracing & Logging
- Structured Logging And Why Text Fails — Full-Stack Observability: Metrics, Tracing & Logging
- Log Levels Done Properly — Full-Stack Observability: Metrics, Tracing & Logging
- Dashboards That Answer Questions — Full-Stack Observability: Metrics, Tracing & Logging
- Alerting On Symptoms Versus Causes — Full-Stack Observability: Metrics, Tracing & Logging
- Alerting Rules And The For Duration — Full-Stack Observability: Metrics, Tracing & Logging
- Service Level Indicators — Full-Stack Observability: Metrics, Tracing & Logging
- Service Level Objectives — Full-Stack Observability: Metrics, Tracing & Logging
- Error Budgets And Burn Rate Alerts — Full-Stack Observability: Metrics, Tracing & Logging
- Ninety Nine Percent Of What? SLIs, SLOs And SLAs — Prometheus Certified Associate (PCA) Exam Prep
- Under A Second, On Average: Histograms — Prometheus Certified Associate (PCA) Exam Prep
- The Export That Failed Quietly: Timestamp Metrics — Prometheus Certified Associate (PCA) Exam Prep
- Thirty Second Checkouts Nobody Saw: Client Libraries — Prometheus Certified Associate (PCA) Exam Prep
- Paged Three Minutes Late: Alerting Rules — Prometheus Certified Associate (PCA) Exam Prep
- Alerting Sprint: Eleven Calls In Three Rounds — Prometheus Certified Associate (PCA) Exam Prep
- Event Correlation Rules, Thresholds & Anomaly Pipelines — Security Operations & Threat Hunting (SOC)
- Adversary Tactics, Techniques & Procedures (TTP) Mapping — Security Operations & Threat Hunting (SOC)
- Recovery & System Validation Testing — Security Operations & Threat Hunting (SOC)
- Post-Incident Review & Root Cause Analysis — Security Operations & Threat Hunting (SOC)
- Observability As A Design Input — Scalable System Design & Distributed Architecture
- Capacity Planning As A Design Input — Scalable System Design & Distributed Architecture
References
- CloudWatch alarm evaluation
- Prometheus: alerting rules
- NIST SP 800-92: Guide to Computer Security Log Management
- Prometheus documentation: Alerting practices