Streaming Pipeline Anomaly Detection Benchmark: Kafka and Flink Telemetry Dataset

Benchmark dataset for ML-based anomaly detection on Apache Kafka and Apache Flink streaming pipeline telemetry. Contains feature files extracted from an e-commerce streaming pipeline deployed on Amazon EKS, with 80 fault injection runs across 8 streaming-specific fault types (broker crash, network partition, disk I/O saturation, consumer lag spike, rebalancing storm, poison pill, checkpoint failure, OOM exhaustion). The dataset includes:- Feature files at 4 window sizes (15s, 30s, 60s, 120s) with 475 features after selection- Fault injection campaign manifests with phase timestamps and labels- Experimental results for 8 ML models (Isolation Forest, OC-SVM, LSTM-AE, Transformer-AE, CNN-AE, LSTM-VAE, Random Forest, Rule-Based) across 5-fold cross-validation- Statistical test results (Friedman, Nemenyi, Wilcoxon, Cliff's delta)- Sensitivity analyses (prevalence, cooldown exclusion, window size, feature category ablation, fusion strategies)- All analysis scripts for full reproducibility Companion paper: "ML-based anomaly detection for real-time data streaming pipeline reliability"