Datasets:

THULab
/

cyb001_sample

license: cc-by-nc-4.0
task_categories:
  - tabular-classification
  - time-series-forecasting
tags:
  - tsfile
  - timeseries
  - cybersecurity
  - network-traffic
  - intrusion-detection
  - synthetic-data
  - anomaly-detection
  - apt
  - c2-beacon
  - time-series
  - format:tsfile
pretty_name: CYB001 — Synthetic Network Traffic (Sample) (TsFile)
size_categories:
  - 1K<n<10K
configs:
  - config_name: default
    data_files:
      - split: flow_events
        path: flow_events.tsfile
      - split: network_flows
        path: network_flows.tsfile
      - split: network_topology
        path: network_topology.tsfile
      - split: session_summary
        path: session_summary.tsfile

CYB001 — Synthetic Network Traffic (Sample) (TsFile)

Apache TsFile version of xpertsystems/cyb001-sample.

Overview

CYB001 — Synthetic Network Traffic Dataset (Sample) — This is a free preview of the full CYB001 — Synthetic Network Traffic Dataset product. It contains roughly 1 / 60th of the full dataset at identical schema, label distribution, and statistical fingerprint, so you can evaluate fit before licensing the full product.

Tables: flow_events, network_flows, network_topology, session_summary
Rows: 16,401 across 4 TsFile table(s).
- flow_events: 5,431 rows — Time is the source timestamp_offset_ms offset in milliseconds.
- network_flows: 9,770 rows, 2024-01-01 01:12:21.832000 → 2024-01-30 23:52:14.441000 (UTC)
- network_topology: 200 rows — Time is synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
- session_summary: 1,000 rows, 2024-01-01 01:10:27 → 2024-01-30 23:47:40 (UTC)

Schema (TsFile structure)

`flow_events.tsfile`

Time (INT64, milliseconds) — the source timestamp_offset_ms offset in milliseconds.
TAG: segment_id — device dimension(s); query one device with WHERE segment_id=....
FIELD (9): flow_id (string), session_id (string), event_type (string), source_port (int64), dest_port (int64), protocol (string), packet_size_bytes (int64), payload_entropy (double), event_severity (string)

`network_flows.tsfile`

Time (INT64, milliseconds) — the source flow_start_timestamp timestamp as INT64 milliseconds.
TAG: segment_id — device dimension(s); query one device with WHERE segment_id=....
FIELD (36): flow_id (string), session_id (string), source_ip_hash (string), destination_ip_hash (string), source_port (int64), dest_port (int64), protocol (string), flow_duration_ms (int64), total_fwd_packets (int64), total_bwd_packets (int64), total_bytes_fwd (int64), total_bytes_bwd (int64), fwd_packet_len_mean (int64), fwd_packet_len_std (int64), bwd_packet_len_mean (int64), bwd_packet_len_std (int64), flow_bytes_per_sec (double), flow_packets_per_sec (double), inter_arrival_time_mean (double), inter_arrival_time_std (double), tcp_flag_syn_count (int64), tcp_flag_ack_count (int64), tcp_flag_fin_count (int64), tcp_flag_rst_count (int64), tcp_flag_psh_count (int64), tcp_flag_urg_count (int64), flow_lifecycle_phase (string), traffic_category (string), attack_subcategory (string), label (string), source_device_type (string), dest_device_type (string), attacker_capability_tier (string), retransmission_flag (int64), fragmentation_flag (int64), protocol_violation_flag (int64)

`network_topology.tsfile`

Time (INT64, milliseconds) — synthetic row index over the stable source order (0 … N-1); the source table has no per-row clock.
TAG: segment_id — device dimension(s); query one device with WHERE segment_id=....
FIELD (17): detector_id (string), segment_type (string), trust_level (double), firewall_policy (string), allowed_protocols (string), avg_concurrent_flows (int64), bandwidth_mbps (double), nat_enabled (int64), ids_coverage (double), qos_policy (string), diurnal_peak_factor (double), feature_space_dim (int64), alert_threshold (double), retraining_cadence_days (int64), ensemble_size (int64), defender_architecture (string), device_count (int64)

`session_summary.tsfile`

Time (INT64, milliseconds) — the source session_start_timestamp timestamp as INT64 milliseconds.
TAG: segment_id — device dimension(s); query one device with WHERE segment_id=....
FIELD (22): session_id (string), source_ip_hash (string), destination_ip_hash (string), session_duration_sec (double), total_flows (int64), total_bytes_transferred (int64), total_packets (int64), dominant_protocol (string), primary_traffic_category (string), label (string), attack_subcategory (string), attacker_capability_tier (string), source_device_type (string), dest_device_type (string), payload_entropy_mean (double), retransmission_rate (double), protocol_violation_count (int64), c2_beacon_flag (int64), exfil_volume_bytes (double), scan_probe_count (int64), lateral_move_flag (int64), session_risk_score (double)

Conversion notes

Each CSV of the sample is converted into its own TsFile table (4 tables), so no table is merged with another and no column is dropped.
Event/telemetry tables use the source simulation clock as Time (INT64 milliseconds): timestep step counters and millisecond offsets are stored as-is, minute-based columns are multiplied by 60,000, and ISO/UTC strings are parsed to epoch milliseconds.
Static inventory/topology/summary tables have no per-row clock; their Time is a synthetic row index (0..N-1) over the stable source order.
Low-cardinality dimensions of each table (campaign/incident/asset/session ids, phases, labels) are declared as TAG columns; query one simulation with e.g. WHERE campaign_id='...'.

Usage

Install the Apache TsFile Python SDK (pip install tsfile) and read a converted file:

from pathlib import Path
from tsfile import TsFileReader

path = Path("flow_events.tsfile")
with TsFileReader(str(path)) as reader:
    schemas = reader.get_all_table_schemas()
    print("tables:", list(schemas))
    table_name = next(iter(schemas))
    table = schemas[table_name]
    columns = [column.get_column_name() for column in table.get_columns()]
    print("columns:", columns)
    field_names = [
        column.get_column_name()
        for column in table.get_columns()
        if column.get_column_name() not in {"Time", "time"}
    ]
    if field_names:
        with reader.query_table(table_name, field_names[:3], batch_size=1024) as result:
            batch = result.read_arrow_batch()
            if batch is not None:
                print(batch.to_pandas().head())

Source & license

Original dataset: xpertsystems/cyb001-sample
Source card title / publisher: CYB001 — Synthetic Network Traffic (Sample)
License: cc-by-nc-4.0 (as declared by the original dataset).