Event-Time Watermark and Late-Data State Simulation Dataset
- Records
- 7200
- Size
- 0.5GB
- Format
- JSON
- Provider
MOBIUSI INC- Updated
- v1.02026-11-04
This dataset provides simulated stream-processing runs with out-of-order events, event-time distributions, watermark strategies, window configurations, and late-data handling rules. It records watermark progression, window completion, pending late events, and aggregation states to show how state changes under different time-semantics configurations. Use it to support state prediction, temporal-semantics modeling, strategy comparison, and testing of stream-processing engines.
Overview
Dataset Insights
Preview
Sample Examples
Schema
Technical Specifications
| Field | Type | Description |
|---|---|---|
| file_name | string | File name |
| simulation_input | object | Raw inputs containing the event stream and the watermark, window, late-data, and simulation-run configurations. |
| aggregation_state | array | Records intermediate streaming aggregation states and updates by window and grouping key. |
| simulation_summary | object | Summarizes event processing, watermark behavior, and final window results for the complete simulation run. |
| pending_late_events | array | Records late events that remain unresolved at the current simulation step and their degree of lateness. |
| watermark_progression | array | Records observed watermarks, progression increments, and states at each simulation processing step. |
| window_completion_state | array | Records the completion ratio, event count, and finalization status for each event-time window. |
Trust
Compliance Statement
Authorization Type
Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use
Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization
No PII, no real company names, simulated scenarios follow industry standards
Compliance System
Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs
FAQ
Frequently Asked Questions
- What does the dataset simulate?
- It covers out-of-order events, event-time distributions, watermark strategies, window configurations, and late-data rules. The run sequences record watermark progression, window completion, pending late events, and aggregation states.
- How is the data structured?
- The dataset includes raw simulation inputs, step-by-step watermark sequences, window completion states, pending late events, aggregation state sequences, and a simulation summary. This structure supports analysis by run step and window state.
- Which training or evaluation tasks can it support?
- It can support state prediction tasks for watermark progression, window completion, late-event handling, and aggregation states, as well as temporal-semantics modeling and strategy comparison for stream-processing engines.
- How can I access the data, and what are the licensing terms?
- Check the dataset product page for available access methods, file specifications, and license terms. Before commercial use or redistribution, verify the applicable license scope and restrictions.
Can't find the data you need?
Post a request and let data providers reach out to you.
Get this Dataset
High-quality data, ready for your next AI breakthrough.
Cite this Work
@dataset{Mobiusi2026,
title={Event-Time Watermark and Late-Data State Simulation Dataset},
author={MOBIUSI INC},
year={2026},
url={https://www.mobiusi.com/datasets/51956ecdc37f37f74fdc446b3b6175e0?dataset_task_cate_id=11},
urldate={2026-10-27},
keywords={event-time dataset,watermark simulation,stream processing data,late-data handling,window state prediction,stream aggregation simulation,time-semantics modeling,stream engine testing},
version={1.0}
}Using this in research? Please cite us.