Design an Anomaly Detection System for Time-Series Data — System Design Interview Practice
Design a system that detects anomalies in high-volume time-series data from metrics, logs, and business KPIs using statistical and ML methods, with adaptive thresholds and root cause hints. Work through the requirements, architecture trade-offs, and an interactive design review.
Requirements and concepts to consider
- Detect point, contextual, and collective anomaliesRequirement
- Handle seasonality and trend in time-seriesRequirement
- Adapt thresholds dynamically based on data patternsRequirement
- Process 100k+ time-series concurrentlyRequirement
- Provide anomaly explanations with correlated signalsRequirement
- Support custom anomaly definitions per metricRequirement
- Minimize alert fatigue with smart groupingRequirement
- Integrate with alerting and incident management systemsRequirement
- mlConcept to explore
- anomaly detectionConcept to explore
- time seriesConcept to explore
- monitoringConcept to explore
- statisticsConcept to explore
- streamingConcept to explore