IcedTea: Efficient and Responsive Time-Travel Debugging in Dataflow Systems

Summary: IcedTea: reverse time-travel debugging for distributed dataflow systems — periodically capture global operator states and allow rollback to any past interaction with stepwise replay. Formalizes the model and applies runtime/space optimizations to support responsive debugging of stateful, ML-heavy dataflows with low overhead. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID: 14434
Venue: VLDB
Year: 2025
Pagerank: 5.093636e-05
Overall Rank: 11,106 | 23.81%
DOI: 10.14778/3712221.3712251

Incoming Non-self Citations Over Time

No non-self incoming citations found for this paper in this database.

Authors

1. Shengquan Ni (University of California Irvine)
2. Yicong Huang (University of California Irvine)
3. Zuozhi Wang (University of California Irvine)
4. Chen Li (University of California Irvine)

BibTeX Citation

@article{ni_vldb25,
        title = {{IcedTea: Efficient and Responsive Time-Travel Debugging in Dataflow Systems}},
        author = {Ni, Shengquan and Huang, Yicong and Wang, Zuozhi and Li, Chen},
        journal = {PVLDB},
        series = {{VLDB} '25},
        volume = {18},
        number = {3},
        pages = {902--914},
        doi = {10.14778/3712221.3712251},
        url = {https://doi.org/10.14778/3712221.3712251},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 1 of 1 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
11,066	ML-Asset Management: Curation, Discovery, and Utilization	2025	VLDB	5.093636e-05

Outgoing Citations (Sorted by Pagerank)

Showing 10 of 10 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
24	Spark SQL: Relational Data Processing in Spark	2015	SIGMOD	0.00054865648
1,617	Titian: Data Provenance Support in Spark	2016	VLDB	0.00010209397
3,885	Apache Tez: A Unifying Framework for Modeling and Building Data Processing Applications	2015	SIGMOD	7.0475239e-05
4,114	Optimizing Machine Learning Inference Queries with Correlative Proxy Models	2022	VLDB	6.8941194e-05
4,273	User-Defined Operators: Efficiently Integrating Custom Algorithms into Modern Databases	2022	VLDB	6.7909197e-05
5,232	A.M.B.R.O.S.I.A: Providing Performant Virtual Resiliency for Distributed Applications	2020	VLDB	6.3067852e-05
6,680	Beyond Analytics: The Evolution of Stream Processing Systems	2020	SIGMOD	5.8049764e-05
8,090	Amber: A Debuggable Dataflow System Based on the Actor Model	2020	VLDB	5.4890583e-05
8,916	Texera: A System for Collaborative and Interactive Data Analytics Using Workflows	2024	VLDB	5.3483178e-05
11,414	Udon: Efficient Debugging of User-Defined Functions in Big Data Systems with Line-by-Line Control	2023	SIGMOD	5.093636e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	2,077	Generating Example Data for Dataflow Programs	2009	SIGMOD
2	834	Debugging Schema Mappings with Routes	2006	VLDB
3	11,443	Fries: Fast and Consistent Runtime Reconfiguration in Dataflow Systems with Transactional Guarantees	2023	VLDB
4	455	Differential dataflow	2013	CIDR
5	9,045	Causality-Guided Adaptive Interventional Debugging	2020	SIGMOD
6	8,185	BugDoc: Algorithms to Debug Computational Processes	2020	SIGMOD
7	361	The Dataflow Model: A Practical Approach to Balancing Correctness, Latency, and Cost in Massive-Scale, Unbounded, Out-of-Order Data Processing	2015	VLDB
8	11,814	Demonstration of Interactive Runtime Debugging of Distributed Dataflows in Texera	2020	VLDB
9	2,196	Spinning Fast Iterative Data Flows	2012	VLDB
10	4,845	Explaining Outputs in Modern Data Analytics	2016	VLDB