Datasets

This is a homepage of network traffic datasets created in CESNET. Our motivation for building datasets is to provide benchmarks for cross comparison of research methods for the tasks of network traffic classification, threat detection, and hardware-accelerated encrypted traffic monitoring.

The following datasets are currently available:

• Dataset of intrusion detection alerts from a sharing platform (2019)
• CESNET-TLS22 – a large dataset for fine-grained classification of TLS services (2022)
• CESNET-QUIC22 – a large one-month QUIC traffic dataset (2022)
• HTTPS
  • HTTPS brute-force dataset with extended network flows (2020)
  • Dataset used for HTTPS traffic classification using packet burst statistics (2021)
• DNS over HTTPS datasets:
  • Dataset used for detecting DNS over HTTPS by Machine Learning (2020)
  • Dataset used for fingerprinting of DNS over HTTPS responses (2020)
  • List of DNS over HTTPS resolvers on the internet (2021)
  • DoH — Real-World (2022)
  • DoH-Gen-F-FGHOQS (2022)
  • DoH-Gen-C-CFGHOQS (2022)
  • DoH-Gen-F-AABBC (2022)
  • DoH-Gen-C-AABBCC (2022)
  • DoH-Gen-C-DDD (2022)
  • DoH-Gen-F-CCDDD (2022)
• Cryptomining
  • CESNET-MINER22: Datasets of Cryptomining Communication (2022)
  • CESNET-MINER22-TS: Periodic Behavior Features of Cryptomining Communication (2023)
• CESNET-AGG23: A descriptive dataset about CESNET2 network characteristics (2023)
• CESNET-USTS23: A benchmark dataset of Unevenly spaced time series from network traffic (2023)
• Network traffic datasets created by Single Flow Time Series Analysis (2023)